Skip to main content

Full text of "Processes of ongoing human evolution"

See other formats

: ' 





edited by 
Gabriel W. Lasker 



Copyright © i960, Wayne State University Press 

Detroit 2, Michigan 

All rights reserved 

Library of Congress Catalog Card Number 6o-iz$66 



Gabeiel W. Laskee 1 

Climate, Culture, and Evolution* 

Paul T. Bakee 3 

Natural Selection, Disease, and Ongoing Human Evolution, as 
Illustrated by the ABO Blood Groups 

Feank B. Livingstone 17 

Metabolic Polymorhpisms and the Role of Infectious Diseases in 
Human Evolution 

Aeno G. Motulsky 28 

Adaptation, Selection, and Plasticity in Ongoing Human Evolution * 

Feedeeick S. Hulse 63 

Migration, Isolation, and Ongoing Human Evolution 

Gabeiel W. Laskee 80 

Irradiation and Human Evolution 

Eaele L. Reynolds 89 


A hundred years ago Charles Darwin published his classic explana- 
tion of evolution through natural selection. The work deals with the 
origin of species, as the title indicates; it is retrospective in outlook, 
explaining past events ; and almost no mention is made of man, except a 
comment that "in the distant future . . . light will be thrown on the 
origin of man and his history/' 

A reading of Darwin's works and that of his successors during the 
last 100 years makes these points clear: First, the theory of natural 
selection applies equally and, indeed, is best understood only by examining 
evolution at subspecific levels. Second, general principles are involved 
which have predictive value, albeit more in respect to the "how" than 
the " what " of human evolution. And third, man is inevitably the most 
interesting subject of evolutionary study. 

The Darwin centennial celebration has included a series of important 


scientific meetings to discuss the status of Darwinism in the thought of 
the last century. In preparing for the present symposium 1 it seemed 
most fit to leave to the earlier meetings, which have been held at London, 
Chicago, and elsewhere, the task of evaluating the contributions of Dar- 
win's work to the thought of the last century. Ample attention has been 
given to the review of studies of natural selection in the various branches 
of science in which it has been applied, and to the developments and 
modifications of the theory which have taken place. 

We thought that a fitting way to close the centennial year would be 
with a small symposium devoted to the question of contemporary human 
evolution. Our topic is the change now going on from generation to 
generation and the focus is on the discussion of man. 

Whatever predictive value the theory of natural selection may have 
can probably best be tested in microevolutionary situations of small scope 
in time and space. Generation-to-generation evolution today is both the 
kind easiest to study accurately and, in a sense, the most important. 
After all, what we believe about the origin of man will make little 
difference, but what we know of ongoing human evolution and what we 
do about it may well determine the future of the species. 

A grant from the Wenner-Gren Foundation for Anthropological 
Eesearch toward the cost of preparing the symposium for publication 
is sincerely appreciated. 

Gabkiel Wakd Lasker 

Wayne State University, 
Detroit, Michigan 

1 A symposium entitled " The Processes of Ongoing Human Evolution as Viewed 
100 Years after the Publication of The Origin of Species " and presented at the 
58th annual meeting of the American Anthropological Association, Mexico City, 
December 28, 1959. The six lectures which comprise the symposium are being 
issued in book form under the title The Processes of Ongoing Human Evolution 
by Wayne State University Press. 


BY P&Ul T. B&KESt 

Pennsylvania State University, 
University Park, Pennsylvania 

IN the almost ninety years since Darwin published " The Descent of 
Man and Selection in Relation to Sex" (1871) many aspects of our 
knowledge about human evolution have changed. So much so that 
Darwin would probably find considerable study necessary to understand 
a symposium on culture or fossil man. However, if he were to join a 
discussion on climate and human evolution, he would be perfectly at 
home with his 1870 concepts. 

In his famous book, he wrote that a naturalist not familiar with the 
races "would be deeply impressed with the fact . . . that the different 
races of men are distributed over the world in the same zoological prov- 
inces, as those inhabited by undoubtedly distinct species and genera of 
mammals." Darwin later discussed the possibility that variations in solar 
radiation in different parts of the world were responsible for the varia- 
tions in the skin colors of races and came to the conclusion : "Although 
with our present knowledge we cannot account for the differences of color 
in the races of man, through any advantage thus gained, or from the 
direct action of climate ; yet we must not quite ignore the latter agency, 
for there is good reason to believe that some inherited effect is thus 

Despite Darwin's feeling that climate was somehow involved in human 
evolution, he could not believe that climate or any other aspect of the 
physical environment had produced the racial variations in our species. 
As so many others he was impressed with the unique potentials of human 
culture to modify the human form. He stressed the role of varying 
concepts of human beauty in determining our form and finally concluded : 
". . . of all the causes which have led to the differences in external 
appearance between the races of man and to a certain extent between 
man and the lower animals, sexual selection has been the most efficient." 

Using statistical tools and data not available in Darwin's time, recent 
studies, have shown that there is a high degree of relationship between 
climatic variables and some of the morphological variables in the human 
species (Roberts, 1953; and Newman, 1953 and 1956). Experimental 


studies have demonstrated that certain of these characteristics provide 
adaptive advantage to the individuals endowed with them. 

Individuals who have a small amount of body fat, great body linearity 
and brunette skin can probably march for substantially longer distances 
in a hot desert than their morphological counterparts (Baker, 1955). 
Individuals with a stocky body build and large deposits of subcutaneous 
fat can sit nude for considerably longer periods in a cool temperature with 
less loss of body heat and less metabolic disturbance than the desert- 
adapted thin man (Baker and Daniels, 1956; LeBlanc, 1954). Experi- 
mental evidence has even shown us that the American Negro who has his 
extremities exposed to below freezing temperatures is much more likely 
to suffer from frostbite than the American White who is exposed to the 
same condition. On the other hand, American Negroes show less devia- 
tion from normal core temperature when they perform work under hot 
wet conditions than do American Whites, even though matched for the 
body linearity and fat, factors which might affect strain levels (Baker, 
1959). Australian aborigines who sleep nude under cold conditions 
apparently have mechanisms of vaso-constriction which permit them to 
conserve body heat and sleep peacefully in a situation where European 
Whites would burn up great quantities of food shivering, while totally 
unable to sleep (Scholander, 1958). 

These are but a few of the newly documented biological adaptations 
of different racial groups to their climatic environment. These examples 
have been selected because they represent those where specific genetic 
inheritance is most clearly indicated. Since this line of investigation 
has become popular in only the last ten years, there is every reason to 
believe that examples of human genetic adaptation to climatic environ- 
ment will multiply rapidly. 

It thus seems that Darwin's belief about the possible role of climate 
in race formation has been verified. However, it is not enough to find 
evidence of climatic adaptation. There remains the much larger question 
of how climatic selection would operate on man's genetic structure to 
produce these adaptations. 

First, there must be the genetic potential and although mutations 
make anything possible, the overlapping of most racial characteristics in 
the various groups of Homo sapiens makes it seem probably that climatic 
selection has worked primarily on the polymorphism of the human 
genetic structure. 

Second, climatic selection must have operated either by modifying 
the fertility of the individuals involved or by the actual death of carriers 


of certain genes before or during their reproductive life span. As of 
now, most evidence points to the latter probability. It is easy to visualize 
the tall skinny Yahgan freezing to death while out in a canoe in a snow 
storm but more probably the family died around a fire suffering from 
malnutrition and consequent disease while their short fat counterparts 
comfortably collected shellfish in the cold water. 

Admittedly this example is speculative. It is so because it relies 
upon placing climatic selection in a cultural context. Yet it is almost 
impossible to imagine climatic selection outside the cultural media. 
Paleontological and archaeological evidence clearly indicates that man 
had already acquired tools and fire before Homo sapiens evolved as a 
species (Washburn, 1959). Most racial characteristics, therefore, evolved 
in groups who had well-developed cultures. Unless the culture form was 
an interactive part of the selective process, climatic selection would not 
have operated as it did on Homo sapiens. Going back to one bit of 
experimental evidence : the Australian aborigine seems to have a genetic 
adaptation to sleeping nude in the cold. This trait could not have been 
selected in a different cultural context. Given clothes, use of the brush 
shelter which he knew how to build, or even lacking the fire which was 
on either side of him during sleep, it is doubtful that the aborigine 
would have been genetically selected for this particular form of vaso- 

To completely reject the concept of climatic selection without cultural 
involvement would be premature, but from anthropology's present theo- 
retical framework it is more accurate always to consider the role of 
culture when trying to formulate a human evolutionary process related 
to climate. At least racial differences in climatic adaptation would seem 
to depend in part on cultural factors. 

Since culture usually forms an essential link in " climatic selection," 
it is important to understand the ways in which climate and culture 
may interact to produce selection. There is, of course, the direct action 
of climate on culture and perhaps the reciprocal, but this belongs more 
properly to cultural evolution and archaeology. Based on the known 
examples of climatic adaptation in man, there appear to be five patterns 
of climate-culture interaction which determine the specific form of 

These five are: 

1. Primary climate-culture interaction in any "climatic selection"; 

2. "Cultural selection" reinforcing "climatic selection"; 


3. " Cultural selection " opposing " climatic selection " ; 

4. Cultural blocks to " climatic selection " ; 

5. Cultural mediation in " climatic selection." 

Kather than define these five categories in detail, it may be more 
enlightening to choose a single body characteristic and show how the 
interaction forms are required to explain the species-wide variation in 
the trait. 


Many of the recent studies relating body morphology to climate have 
emphasized the surface-area to weight ratio as an important trait in 
climatic adaptation. Coon, Garn and Birdsell (1950) pointed out that 
larger surface area per unit of weight might result in a greater loss of 
heat from the body to the surrounding environment. They, therefore, 
felt that the attenuation of the Nilotic Negro formed a body adaptation 
to hot desert conditions. This form was contrasted to the short square 
body build of the Eskimo, who was presumably cold adapted. Eoberts 
(1953) later presented quantified data on the relationship between the 
separate entities of weight and stature to climate, but it remained for 
Schreider (1950 and 1951) actually to combine this data into a surface- 
area over weight ratio and correlate it to the climatic environment of 
different populations. 

Despite reasonably conclusive evidence of a human SA/W relationship 
to climate it cannot be, thereby, concluded that this is a racial variation 
which has developed as a climatic adaptation. First it must be experi-t 
mentally shown that variations in this trait confer specific climatic 
adaptation. Second, it must be shown that the degree of racial variation 
found for this ratio is at least in part indicative of group differences in * 
genetic structure. 


In contradiction to the original hypothesis put forth by Coon, Garn 
and Birdsell, experimental work has thus far failed to show that a high 
SA/W ratio provides any great physiological advantage to hot-desert 
dwellers (Baker, 1955; and Adolph, 1947). Beyond lowering water 
requirements because of a small body size, a high ratio has no appreciable 
effect on man's desert heat tolerance. Actually this could have been 
anticipated from a detailed knowledge of human physiology. The human 
body depends primarily upon the cooling derived from sweat evaporation 


for maintaining thermal homeostasis in a hot desert. The hot dry air 
of the desert has enormous evaporative power and is apparently capable 
of evaporating sweat much more rapidly than the human body can 
produce it. Since the sweat production of the active man is related 
more closely to his fat-free body mass than it is to his surface area, the 
total cooling per unit of weight would be predicted to be very similar 
for men of quite different surface areas. 

Continuing on a theoretical basis, a high SA/W ratio may provide 
decided advantage to the man who must do physical exercise under hot 
wet climatic conditions. When the air has a high moisture content 
combined with high temperatures, such as that in tropical forest regions, 
it no longer has the capacity to evaporate all the sweat produced 
by an active man. Under these conditions a significant proportion of 
the sweat will form water droplets and run off the man, providing no 
body cooling. Thus in the hot wet climate, the total surface area over 
which a given sweat production is spread will govern the amount of 
cooling derived. Since the amount of sweat produced by a man is 
governed by his weight of fat-free mass then, with activity held constant, 
increases in SA/W ratio should lower the heat strain on a man in the 

Unfortunately this theoretical prediction is without experimental 
verification simply because no one has yet performed the experiment. 
The only direct evidence which can be offered at this time is the common 
observation that large men, who have low SA/W ratio, appear to produce 
a greater proportion of unevaporated sweat when they work hard in a 
hot wet environment. 

Only in the cold is Coon, Garn and BirdselFs hypothesis fully sub- 
stantiated by both the theoretical and experimental evidence. If the vaso- 
constriction of different groups of men is the same, and if the amount 
of subcutaneous fat is the same, it is a thermodynamic law that heat loss 
is in direct relation to surface area. This rather dogmatic statement 
may be made because man's heat loss in the cold is a simple process of 
conductance and radiation. The shape factor may have some slight 
effect but it seems safe to assume that the morphological characteristics 
of men with low SA/W ratios would not be in the direction of more 
pointed or angular surfaces. Since the surface area controls the heat 
loss, and fat-free body weight (with other factors held constant) controls 
the heat production, it becomes axiomatic to state that for nude men, 
the lower the SA/W ratio the lower the heat loss per unit weight. 

Experimental studies have shown that, for American Whites, variation 


in subcutaneous fat is probably a more significant factor in cool climate 
homeostasis than SA/W ratio variations (Baker and Daniels, 1956; 
Baker, 1959 ; LeBlanc, 1954) . However, when subcutaneous fat was held 
constant, individuals with lower ratios maintained higher core tempera- 
tures at lower weight-adjusted metabolic rates (Baker and Daniels, 1956; 
Baker, 1959). 

From the present theoretical and experimental evidence it appears 
that at least under hot, wet and cold conditions, the first requirement 
has been fulfilled and the correlations of SA/W to climate mean some 
kind of functional advantage to the groups involved. 


The second requirement was a genotypic involvement in the previously 
mentioned variations in SA/W ratio. Before questions of genotype can 
be discussed, the rather complex SA/W ratio or index must be broken 
down into its parts and morphological determinants. Surface-area to 
weight ratios as reported by almost all investigators are no more than a 
special function of height and weight. Since actual surface area is an ex- 
tremely time-consuming measurement, most students rely on DuBois' 
(1936) original study of 8 men or perhaps the later mathematical refine- 
ments (Sendroy and Cecchini, 1954). These formulae are reasonably 
efficient when applied to White Europeans, and Ehodal (1952) reports 
that they also apply to Eskimos. Despite these studies it is hard to believe 
that the simple measurement of height and weight will provide equally 
accurate estimates of surface area for the short-bodied, long-legged 
Nilotic and long-bodied, short-legged Mongoloid. It seems to me quite 
probable that the Nilotic has a greater surface area than estimated from 
his height and weight while the Mongoloid's is lower than estimated. 

The most direct factor involved in the ratio is weight. Without 
becoming too deeply involved in the reasons, one can point out that as 
body weight goes up the SA/W ratio goes down. Simple solid geometry 
shows that going from a one-inch cube of wood to a two-inch cube in- 
creases its weight eightfold but increases its surface area only fourfold. 
While human weight to surface area does not progress in such a simple 
manner, surface area does not proportionally increase with weight. This 
means that a 160 lb. man always has a lower SA/W ratio than a 120 lb. 
man, provided his stature or body proportions are not pathologically 

The genetic involvement of the ratio may therefore be judged pri- 


marily by the extent to which body weight is genetically determined, 
secondarily by the genetic determination of stature, and to some unknown 
extent by the genetic involvement in body-trunk to leg-length ratios. 

Quite obviously both the weight and stature of groups of men are 
variable depending on the action of the physical and cultural environ- 
ment in which the phenotype develops. In the terms Hulse uses in his 
contribution to this symposium, height and weight are among the most 
plastic of human characteristics. Nevertheless, no one has shown or 
suggested that a group of African Pygmies living in American culture, 
even for several generations, would acquire the height and weight of 
European-derived Americans. Although there may also be some changes 
in trunk to leg ratios, the American Negro still has a longer leg in 
relation to trunk than the American White (Hooton, 1959). The exact 
proportion of plasticity vs. polymorphism in the racial distribution of the 
SA/W ratio cannot be quantified. But it seems safe to state that at 
least part of the group variation in this ratio is related to differences 
in gene frequencies. 


Our ratio has now fulfilled the two essential criteria and appears to be 
an example of a body characteristic which has been climatically selected. 
This brings up the much more difficult question of how the selection 
took place. It is in the potential modes of selection that it becomes 
obvious culture can never be discounted. 

To begin with basic assumptions : it is assumed that all of the recorded 
group genetic variation in the SA/W ratio is the result of plasticity and 
polymorphism. That is, given the time and selective pressure, which 
existed during past epochs, Pigmies would have acquired the stature of 
Europeans or Europeans the stature of Pigmies, had their roles been 
reversed. This assumption obviates the necessity for reconstructing the 
size and shape of a proto-Pigmy or proto-Mlot since, whatever the size 
and shape of the ancestral group, if it had a Homo sapiens type of size 
polymorphism, the group would have (with appropriate selection) evolved 
to its present SA/W ratio. 

In the previous section, the potential advantage of a low or high 
SA/W ratio was noted in relationship to very particular situations. 
Thus, in the hot wet tropics an advantage might accrue to an active 
individual if he had a high ratio. The actual degree of advantage would 
increase as the activity increased. For two men playing checkers in the 
shade of a tree, it is extremely doubtful that any difference in their 



core temperature or performance could be found, whatever the individual 
difference in SA/W ratio. On the other hand, if their culture demanded 
hard work, the men with a low ratio would probably suffer from a sub- 
stantially higher core temperature, might therefore have a lower per- 
formance potential (Fine and Gaydos, 1959) and, if the work was hard 
enough, would either quit or die of heatstroke. When the culture 


■< 1 < , >. 







Fig. 1. Graphic Depiction of the Climatic Selection Process: Climatic 
and Behavioral Factors on the Top Line Determine the Position of a Given 
Population Distribution of Surface Area Per Unit of Weight (SA/W) in 
Relation to the Selective Phenomena Shown Below. 

requires high metabolic output many modes of selection are possible — 
social selection or inability to obtain enough food or even death of the 
" unfit." However, the important point is that culture must be involved 
and unless the culture form required considerable activity, "climatic 
selection " could not operate. 

The accompanying illustration is an attempt to indicate the primary 
selective process. Since SA/W is believed to be normally distributed, it 
has been depicted by the normal curve (figure 1). As the combined 
climatic and cultural factors shown at the top act to bring a population 
into heat or cold stress areas the individuals at one extreme of the dis- 


tribution will be more frequently and drastically stressed than those at 
the opposite end. However, other factors distributed by chance probably 
make the individual SA/W selection group another normal curve with 
its mean near the extreme being stressed. 

Although this figure may not be necessary to explain the process, 
depicting the selection for normally distributed traits in this manner leads 
to some interesting questions. For example, it suggests that groups 
living in hot wet areas may have SA/W ratio distributions which are 
skewed to the left while those in cold stress areas show skewness to the 
right. To the author's knowledge, no one has investigated the problem. 


The low correlations between SA/W ratios and climatic character- 
istics demonstrate quite conclusively that climatic selection does not 
alone determine the ratio of any given group. A number of other factors 
of the physical environment are undoubtedly involved, but the problem 
also remains whether any purely cultural factors may be operative. 

One of the more obvious cultural factors is the standard for mate 
selection. Eecent studies have shown that in most non- Western societies 
virtually everyone has some opportunity to mate. Nevertheless, the fre- 
quency of mating opportunities varies with how closely the individual 
approaches the cultural standards of beauty. Ford and Beach (1952) 
showed that standards of beauty vary considerably from one society to 
the next and it is easy to see how variation in body size preference might 
act to shift population SA/W ratios in the direction of the size preferred. 
Whether these culturally determined ratio shifts were adaptive to the 
climate would depend entirely on the group's climatic habitat. 

Ford and Beach report that the Chukchee prefer females who have 
a plump body. For this group the body form preference may have acted 
to improve their climatic adaptation by lowering their ratio in a cold 
climate. On the other hand, these authors note that the Maricopa also 
prefer plump female mates. This would lead to a poor climatic adjust- 
ment in these hot-desert dwellers. Southwest desert Indians are notable 
exceptions to the general relationship between environmental temperature 
and body build (Baker, 1958), suggesting that sexual selection may over- 
ride climatic selection in some instances. 

As mentioned earlier, climate may have some direct relationship to 
cultural traits but if ignored these examples show how climate-free 
cultural factors may act to augment or detract from the morphological 
adaptation of man to his climatic niche. 



It might be thought that the cultural component of some climatic 
selections could be ignored on all but a theoretical level since most 
cultures demand outdoor activity of their people. The major exceptions 
are obviously the very recent "Westernized" cultures which provide 
sedentary indoor activity. Although statistical evidence is lacking, it is 
doubtful whether anyone would suggest that Americans in Montana or 
Southern Arizona are now being climatically selected at a genetic level. 
However, there are other earlier culture forms where at least some seg- 
ments of the population were culturally shielded from climatic selection. 
Any society which has class or caste, and this includes almost all agri- 
cultural societies, reduces the climatic exposure of the upper strata. 
Inbreeding upper classes have been reported larger in body size for almost 
all societies (Coon, Garn and Birdsell, 1950). This difference is usually 
ascribed to better nutrition, and diet is undoubtedly important. How- 
ever, there are other factors involved. In places like India and central 
Africa, the larger body size of the upper class is ascribed to the social 
dominance of an invading large-size race (Risley, 1915). If this formed 
a complete explanation, it is peculiar that the reverse case of small 
upper-class people has not been reported somewhere in the world. 

It is probable that there is no climatic selection for larger body size 
in the upper class but neither is there climatic selection against this 
large size. In hot climates the lack of essential high activity for most 
upper-strata members breaks the hypothetical selection process. What- 
ever their SA/W ratios, these individuals have equal opportunity of 
reproducing and will not be subject to death or reduced nutrition because 
of their heat tolerance. It may be that much of the class or caste 
difference in the SA/W ratios is simply the result of phenotypic plas- 
ticity, but if any of it is polymorphism this difference may be partially 
the end result of a failure in the cultural selection link. Again, without 
any clear evidence to support the hypothesis, it might be suggested that 
the high SA/W ratios in tropical populations may be maintained in a 
manner similar to the high incidence of sicklemia in the African malarial 
areas. That is, as long as climatic selection for heat tolerance is con- 
tinuing, there is a high ratio, but when the selective process is disrupted, 
other selective phenomena produce a rapid rise to a larger body size and 
consequently lower SA/W ratio. If there were evidence showing a 
smaller body size for the upper strata of cold-area societies a stronger 
argument could be provided for this hypothesis. 



Failure of the culture link can, therefore, block many forms of cli- 
matic selection, but in doing so it may also start a new climate-linked 
selective process in operation. As long as men had no clothing they 
could not survive in climates which had temperatures below freezing. 
With the acquisition of clothing, the selection for SA/W was greatly 
reduced since man could now utilize the much more effective adaptive 
trait of his fellow mammals — their fur. With fur clothing man pene- 
trated far into the zones of the world with below freezing temperatures. 
Although clothing effectively helps to maintain core temperatures, it fails 
adequately to protect man's extremities at very low temperatures. Thus a 
new type of selection came into operation and morphological plus func- 
tional mechanisms which would reduce susceptibility to frostbite were 
now selected. 

When large differences in the incidence of frostbite occurred between 
Negro and White soldiers in the two World Wars, they were attributed to 
cultural differences. However, during the Korean conflict a large and 
very careful epidemiological study was performed. In this study all the 
potential cultural factors in frostbite, down to the number of changes of 
socks were studied. Even with all the training and behavioral factors 
controlled, American Negroes had a much higher incidence of frostbite. 
Orr and Eanier (1951) estimated that, with all other known factors 
controlled, American Negroes born in the North had five times the 
incidence of frostbite of American Whites also born in the North. In 
fact, the racial difference was about the most important etiological factor 

These results prompted climatic physiologists to search for a genetic 
factor which would account for the difference. On the basis of earlier 
Japanese work (Yoshimura and Iida, 1950 and 1952), Meehan (1955) 
studied the temperature changes of the finger when men otherwise ther- 
mally equilibrated put their fingers in ice water. In a large group of 
Negro and White soldiers he found the Whites responded by temperature 
cycling much more frequently than the Negroes. A small group of 
Eskimos showed an even stronger cycling response than Whites. Since 
a pair of identical twins in the study had virtually identical responses, 
he concluded that the differences were probably genetic. 

Two subsequent studies by other investigators have reconfirmed the 
Negro-Whites differences found by Meehan. Eennie and Adams (1957) 
cite evidence to the effect that these differences would be related to frost- 


bite susceptibility. In the most recent study Iampietro et al. (1959) 
show that in a group of 17 White admixed American Negroes and 18 
Whites the lowest 25% of responders were all Negroes while the top 25% 
were all Whites. 

Even though additional study of the temperature " hunting " response 
is indicated for other racial groups, there is enough evidence to indicate 
a genetic adaptation. Yet, we know that without well made clothing 
men could not have survived in an environment where frostbite could 
have occurred. 

Thus, the development of a technology which reduced or perhaps 
stopped the selection of low SA/W ratios created a new environment 
where culture and climate acted together to produce a new and effective 
selective force. 


When Darwin tried to explain the morphological variations in man 
he noted that although there were some obvious correlates between cli- 
matic zones and human morphology it was equally obvious that cultural 
differences could have produced great variation. We have continued to 
separate climate and culture when considering the factors in racial 
evolution. A close examination of the problem suggests that both climate 
and culture are involved in any form of climatic selection. Thus, it 
becomes important to understand the ways in which climate and culture 
can interact to effect genetic selection. 

Eive forms of interaction have been suggested, but more may become 
apparent as new knowledge on the adaptive value of racial differences is 
attained. These five forms treat climate and culture as independent 
variables and ignore the possible direct relationship between these divi- 
sions of man's environment. This discussion does not stress the role of 
other physical environmental factors in determining race differences, but 
it should be remembered that probably most racial differences are the 
product of multiple selective factors and even climate and culture 
together will not explain the total interracial variability in morphological 

To demonstrate the modes of selection created by the five forms of 
climate-culture interaction, group variations in surface area over weight 
are analyzed in the interaction framework. An examination of this 
morphological trait shows : 


(1) a relationship to climatic zones; 

(2) that the group variations have adaptive value; 

(3) that some of the variation reflects genetic polymorphism. 

From this evidence it is concluded that "climatic selection" has acted 
to produce some of the racial differences in this ratio. Consideration of 
the selective modes shows how the selection might have occurred and also 
some of the reasons why there is not a perfect relationship between 
climatic zone and the surface area over weight ratio. 


Adolph, E. F. 1947 Physiology of Man in the Desert. Interscience Publishers, 

Inc., New York. 
Baker, P. T. 1955 Relationship of Heat Stress to Gross Morphology. Env. 

Prot. Div. Technical Rept. EP-7, QM R&D Center. 

1958a Racial differences in heat tolerance. Am. J. Phys. Anthrop., 16 : * 


1958b The biological adaptation of man to hot deserts. Am. Naturalist, V 

42: 337-357. 

1959 American Negro- White differences in the thermal insulative aspects 

of body fat. Human Biol., 31: 316-324. 

Baker, P. T., and F. Daniels, Jr. 1956 Relationship between skinfold thickness 

and body cooling for two hours at 15°C. J. Applied Physiol., 8: 409-416. 
Coon, C. S., S. M. Garn and J. B. Birdsell 1950 Races: A Study of the 

Problems of Race Formation in Man. Charles C. Thomas, Springfield, 

Darwin, Charles 1871 The Descent of Man and Selection in Relation to Sex. 

Murrey, London. 
DuBois, D. 1936 Basal Metabolism in Health and Disease. Lea and Febiger, 

Fine, B. J., and H. F. Gaydos 1959 The relationship between individual per- 
sonality variables and body temperature response patterns in the cold. 

Env. Prot. Res. Div. Technical Rept. EP-106, QMR&E Command. 
Ford, C. S., and F. A. Beach 1952 Patterns of Sexual Behavior. Harper & 

Brothers, New York. 
Hooton, E. A. 1959 Body build in a sample of the United States Army. Env. 

Prot. Res. Div. Technical Report EP-102. QM R&E Command. 
Lampietro, P. F., R. F. Goodman, E. R. Buskirk, and D. E. Bass 1959 Response 

of Negro and White males to cold. J. Applied Physiol., 14 : 798-800. 
LeBlanc, J. 1954 Subcutaneous fat and skin temperature. Canad. J. Biol, and 

Physiol. 32: 407-417. 
Meehan, J. P. 1955 Individual and racial variations in a vascular response to 

a cold stimulus. Military Medicine, 116 : 330-334. 


J Newman, M. T. 1953 The application of ecological rules to the racial anthro- 
pology of the aboriginal New World. Am. Anthrop., 55: 311-327. 
•* 1956 Adaptation of man to cold climates. Evolution, 10: 101-105. 

Orb, K. D., and D. C. Fanier 1951 Cold Injuries in Korea during the Winter 
of 1950-1951. Army Medical Research Laboratory, Fort Knox, Kentucky. 

Rennie, D. W., and T. Adams 1957 Comparative thermoregulatory responses 
of Negroes and White persons to acute cold stress. J. Applied Physiol., 11 : 

Risley, H. 1915 The People of India. W. Thacker & Co., London. 
'Roberts, D. F. 1953 Body weight, race and climate. Am. J. Phys. Anthrop., 11 : 

Rodahl, K. 1952 The body surface area of Eskimos as determined by the linear 
and the height-weight formulas. Am. J. Phys. Anthrop., 10: 419-426. 

Scholander, P. F., H. T. Hammel, J. S. Hart, D. H. LeMessurier, and J. Stem 
1958 Cold adaptation in Australian Aborigines. J. Applied Physiol., 13: 

Schreider, E. 1950 Geographical distribution of the body-weight/body-surface 
ratio. Nature, 165: 286. 

1951 Anatomical factors of body-heat regulation. Nature, 167: 823-824. 

Sendroy, J., and L. P. Cecchini 1954 Determination of human body surface 

area from height and weight. J. Applied Physiol., 7: 1-12. 
Washbubn, S. L. 1959 Speculations on the interrelations of the history of tools 

and biological evolution. Human Biol., 31: 21-31. 
Yoshimura, H., and T. Iida 1950 Studies on the reactivity of skin vessels to 

extreme cold. Part I. A point test of the resistance against frostbite. 

Japanese J. Physiol., 1 : 147-159. 

1952 Studies on the reactivity of skin vessels to extreme cold. Part II. 

Factors governing the individual differences of the reactivity or the resis- 
tance against frostbite. Japanese J. Physiol., 2: 177-185. 



Department of Human Genetics 
University of Michigan, Ann Arbor * 

SINCE Darwin first proposed his theory of evolution one hundred 
years ago, it has been amply proven and hence has become the 
cornerstone of modern biology. Darwin's major contribution was his 
emphasis on the importance of natural selection in the evolutionary 
process. Although most of the advances in physical anthropology in the 
last one hundred years have been due to the application of Darwin's 
theory to the human species, it seems to me that the tremendous implica- 
tions of the great importance of natural selection have yet to be realized. 
Instead, the similarities and differences between living human populations 
have been explained by the concepts of pure race and migration, or more 
recently by random gene drift. For example, the similarities among the 
Pygmy populations which are scattered from Central Africa to Australia 
have been considered to be due as much to their common ancestry as to 
the action of natural selection despite the fact that all these populations 
lived as hunters in tropical rain forest, and there has been much specu- 
lation about the location of the original " pure " Pygmy population and 
what migrations have occurred. 

Now we are all aware that evolution is a complex process, and 
migration and gene drift have obviously played a role in determining 
the distributions of many phenotypic and genotypic characteristics among 
human populations. But what I would like to emphasize is that natural 
selection, the prime mover, has not even been considered in much of 
anthropological theorizing, particularly with regard to the blood groups. 
It is a well-known fact that there are differences in fitness among the Eh 
genotypes, but the sole effect of this fact on anthropological thinking has 
been that the distribution of the Eh genes in Europe has been "ex- 
plained" by the mixture of "pure" races and not impure ones. In 

* Present address : Department of Anthropology, University of Michigan. 


recent years, however, there are increasing signs of the belated recognition 
of the importance of natural selection, and perhaps there is no better 
illustration of the changes in anthropological thinking than the history of 
blood group research. 

After the discovery by the Hirszfelds that ethnic groups differed in 
their frequencies of the ABO genes, great efforts were made in the 1920's 
and 1930's to sample the world's populations, and the general features 
of the world distribution of the ABO blood groups were known when 
Boyd (1939) compiled his famous summary. At the same time, or 
perhaps a little earlier, there was a similar burst of investigations on the 
association of the blood groups with many different diseases. All types 
of disease including infectious, degenerative, malignant, and mental, 
were investigated, and some significant associations were discovered — 
including one between cancer and blood group A by Johannsen in 1927. 
When considered as a whole, however, these investigations were certainly 
equivocal, and for just about every disease there were contradictory 
results, some of which even reported significant associations in opposite 
directions for the same disease. As more careful work and larger samples 
tended to disprove some of these associations {e.g., Parr, 1930), the 
tendency arose to discard them entirely. Thus, the idea became prevalent 
that the blood groups were not significantly associated with any disease 
and hence were "non-adaptive." 

I think that hindsight now indicates that this was a classic example 
of a too common phenomenon in science, "throwing out the baby with 
the bath." But the reasons for this are not difficult to see. The science 
of statistics was in its infancy, and the chi-square test was practically 
unknown. Hence in most of the investigations on the association of the 
blood groups with disease there was no attempt to test the significance 
of the association. Whether or not an association was claimed in many 
of these investigations thus seemed to depend for the most part on the 
attitude of the investigator. In addition the science of population 
genetics had yet to be developed, so that the implications of the world 
distribution of the ABO blood groups, which have been explored by 
Brues (1954) and others, could not be envisaged. And finally many of 
the investigations claimed too much genetic significance for these asso- 
ciations and hence incurred the wrath of the geneticists. The findings 
of an association between the blood groups and a particular disease was 
claimed to indicate genetic linkage between susceptibility to the disease 
and the blood group genes (Hirszfeld, Hirszfeld and Brokman, 1924). 
Thus, when the geneticists disposed of this faulty genetic reasoning in 


an easy fashion (Snyder, 1926), there was a tendency to think that the 
whole problem had been disposed of. However, the associations remained 
and were forgotten for almost twenty-five years. 

The discovery of erythroblastosis fetalis due to maternal-fetal incom- 
patibility of the Eh and ABO blood groups was the first clear evidence 
for selection operating on the blood groups. And shortly thereafter the 
creeping realization began that other selection must be involved in order 
to account for the world distribution. And now the pendulum has 
completed its swing and we are back to testing for associations between 
the blood groups and every imaginable disease. However, typhoid fever, 
diphtheria, scarlet fever, and many other serious diseases of thirty years 
ago are no longer present in any intensity, so that the possibility of 
examining their relationship to the blood groups is many times more 
difficult today. Thus, we are left with stomach cancer, ulcers, and other 
degenerative diseases of modern man. The selective effect through the 
ages of such degenerative diseases would not appear to have been very 

The precise manner by which natural selection has operated on the 
ABO and other blood groups will have to be demonstrated by proven 
associations between them and various diseases, and perhaps later by 
experimental work on the physiological basis of the interaction. How- 
ever, estimates of the amount of selection can be obtained by analysis of 
the distribution of the blood groups as Brues (1954) has done, and at 
the same time insight can be gained into the inter-relationships between 
the factors which can contribute to gene frequency change. The rest of 
this paper will attempt to analyze the distribution of the blood groups 
in West Africa by the methods of population genetics. 

Figure 1 shows the distribution of the frequencies of the A and B 
genes among all the populations of Niger-Congo speakers in West Africa 
for which there are data, with the exception of some early studies which 
differ significantly from later studies of the same people. I have used 
the populations of West Africa because of my own interests, but I am 
sure there are better data for this type of analysis from other parts of 
the world. It can be seen that there is a very considerable clustering 
around the mean values, which appear to be almost identical for A and 
B, and that this clustering approximates a somewhat leptokurtic normal 
curve. This curve is remarkably like that which would be expected if 
constant pressures were tending to equate the frequencies and gene drift 
were tending to separate them. 



The fact that there are single peaks to these curves seems to indicate 
that the tribes of West Africa are rather homogeneous with respect to the 
ABO blood groups. Other aspects of the distribution also seem to indicate 
this. While for the world distribution of the blood groups, Brues (1954) 
found evidence of some interaction between the A and B frequencies, 
for West Africa there is no correlation between these frequencies and 
they seem quite independent. There also seem to be no significant clines 
for the ABO blood groups in West Africa. When the tribes are divided 
into linguistic sub-families and the frequencies tested by an analysis of 
variance, the mean frequencies of the sub-families do not differ signifi- 

4 8 12 16 20 24 

4 8 12 16 20 24 

Fig. 1. The Distribution of the Frequencies of the A and B Genes 
in West Africa. 

cantly; so that there appears to be no relationship between linguistic or 
ethnic affiliation and the blood groups. Thus, the assumption that these 
distributions represent steady states with gene drift tending to separate 
the frequencies and constant pressures tending to equate them seems 
fairly well fulfilled. 

The constant pressures which are counteracting the effects of gene 
drift could be due to any of the three factors, selection, gene flow, and 
mutation. However, since there is no evidence for mutation at the ABO 
locus and the mutation rates required to account for these distributions 
would be approximately 1 in 400 and thus higher than any known rates, 
the role of mutation appears to be quite small in determining these 
distributions and will be disregarded. The problem then becomes one of 
estimating the relative effects of gene flow and selection. To do this 
it is necessary to make two further assumptions. First, I have assumed 


that the average size of the isolate in West Africa is 1000. Actually 
I think it is smaller than this figure, but on the other hand most of the 
data on gene frequencies is recorded by tribal unit, and the tribes of 
West Africa, which average perhaps 10,000 to 100,000, contain several 
breeding isolates. Thus, the figure of 1000 is a happy medium between 
these two. For all these calculations I have also used Wright's (1951) 
island model and not the more complex neighborhood model. For 
an area the size of West Africa, the neighborhood model would un- 
doubtedly be more appropriate but the calculations with this model are 

The amount of gene drift which is possible among a group of popu- 
lations is dependent on N, the average effective size of the populations, 
and to a certain extent on the gene frequency. The amount of gene 
flow is of course dependent on the interchange of people or whole sets 
of genes between the populations. Thus, for all genes the amounts of 
gene drift and gene flow which can occur are the same and not dependent 
on any characteristic of the gene. But the degree to which they can 
influence the frequency of any gene is dependent on the intensity of the 
other factors, mutation and selection, on the gene. If for a group of 
populations in the steady state we assume that for all genes the effects 
of gene drift are entirely balanced by gene flow, then the minimum 
value which we obtained for gene flow would be closest to the actual value 
and hence is the best estimate. This is because all these values include 
the effects of both selection and mutation, and the minimum value is 
the one for which these are a minimum. When the effects of gene drift 
are entirely balanced by gene flow, the variance of the steady-state gene 

frequency distribution is : <r 2 = j^= — — ^-, where N is the effective size 

of the populations and m the amount of gene flow. By equating this 
value to the actual variances of several genes we can get estimates of m, 
and these estimates for West Africa are shown in table 1. 

It will be noted that the values of the variance are lower and hence 
the m's are higher for the A and B genes. This means that the constant 
pressures which are tending to equate the frequencies are greater for 
A and B, which seems in turn to indicate more selection. While the 
variances for the A and B genes are not significantly different, both are 
significantly different from the other three, which are also not signifi- 
cantly different from one another. From these calculations it appears 
that the maximum value for m in West Africa is .01, but it must be 


Estimates of the amount of gene flow (m) among the populations of West Africa 




GENE FLOW ( m ) 




( ASSUMING N = 1000) 



55 .001015 




55 .001351 




27 .004611 




29 .005694 


Rh-(incl.D u ) 


46 .003894 

. Qil — q) 
a — 47Vm + 1 

q(l — q) 


"'"*'" "" 42V 

where N is the effective size of the populations; m, the amount of 

gene flow; and 

q, the average 

gene frequency of the populations. 

remembered that this value is obtained by assuming that there is little 
selection acting on the Eh and MN genes. 

With this estimate of gene flow we can now calculate the amount of 
selection which is acting on the ABO genes. The steady-state equation 
which includes the effects of gene flow and selection is: 

<t>(q)=C(Wy» q u-i(l- q )V-i ; 

where W is the average fitness, U = 4JSfmq, and V = 4lVra (1 — q) . The 
constant in this equation cannot be evaluated, since neither I nor my 
mathematician associates have been able to integrate it. However, we 
can get an approximate solution of the fitness values by setting <j>(q) 
evaluated at .10 equal to 3#(g) evaluated at .15, from which equation 
the constants cancel out. By treating the locus as a two-allele system, 
A and not A, we can solve for the fitness of the A A homozygote. Since 
the shape of the curves for the A and B genes are so similar and their 
means and variances not significantly different, I have set the fitness of 
the BB homozygote equal to that of the AA homozygote. Then by 
examining in a similar way the distribution of the proportion of the A 
and B genes which are A, the fitness of the AB heterozygote can be 
estimated. Armed with these estimates, and setting the fitness of the 00 
homozygote equal to 1, we can obtain estimates for the fitness of the AO 


and BO heterozygotes by using the fact that W, the average fitness, is a 

maximum at equilibrium and hence that - — = at that point. Table 2 

shows the estimates of the fitnesses of the various genotypes of the ABO 
blood groups which are obtained from these calculations. 

The fitness of the various genotypes of the ABO locus 














Derived from: 

0(g) =C{W 

2^P-l (1 _ 



W is the average fitness, U = 

iNmq, and V 

= 4tfm(l- 


We can now test these fitness values by the procedure outlined by 
Kimura (1956) and find that they represent a stable polymorphism and 
thus that the frequency of any population will tend to return to the 
equilibrium value when it is disturbed due to random gene drift. Despite 
the many assumptions which were necessary to calculate these values, 
I think they do show the approximate amount of selection which is 
operating on the ABO system, or at least they do indicate the propor- 
tionally greater amount of selection needed for this locus than for the 
other blood group loci. In this regard their order of magnitude is about 
the same as Brues' (1954) coefficients for the entire world distribution, 
although her values do not represent a stable polymorphism since the B 
gene would tend to disappear. These calculations also indicate the type 
of data that is needed and some of the problems which confront attempts 
to determine the operation of selection at the ABO locus. With the 
development of the mathematical methods of population genetics, data 
from which the variables, gene flow and effective population size, can be 
estimated directly are the major need, although studies like that by 
Cavalli-Sforza (1958) are beginning to fill the gap. A technical advance 
which has almost become a necessity is a method of distinguishing AA 
and BB homozygotes from AO and BO heterozygotes respectively, since 
present typing methods lump individuals with the highest and lowest 
fitnesses. With this innovation age trend analysis may well indicate 
significant changes. 


If selection is playing a major role in maintaining the ABO blood 
group polymorphism, then the question arises as to the nature of this 
selection. The evidence at present points to disease as the method by 
which selection operates. Most of the recent evidence has concerned 
malignant or degenerative diseases, but I think a good case can be made 
for infectious diseases — particularly since these act in the early years of 
life and thus have a greater selective effect. The blood groups result 
in differences in the antigenic structure of the human organism, and 
since the antigen-antibody system of the organism is its chief defense 
against infectious disease, differences in this system may well lead to 
different responses to some diseases. For higher groups of animals we 
know that the parasites of one genus are often unable to parasitize another 
genus because they are quickly identified as foreign antigens and 
destroyed. Perhaps for smaller antigenic differences the same principle 
holds. For example, blood group A is antigenically similar to several 
pneumococci; hence persons with anti-A may resist pneumococci more 
strongly than persons with blood group A, who would have difficulty pro- 
ducing an antibody to their own system. It then follows that persons 
with blood group A would be more susceptible to some types of pneu- 
monia and there is some evidence that this is the case (Struthers, 1951). 
The innumerable investigations which have indicated antigenic similarity 
between the blood groups and a great many organisms, which range in 
size from rickettsia to helminths, may also indicate different responses 
among the blood groups to the diseases caused by these organisms 
(Koppisch and Oliver-Gonzalez, 1959). Some of these are ascaris, 
typhoid, streptococci, staphylococci, Shigella, Proteus, and typhus, in 
addition to pneumococci, There is also evidence for other types of inter- 
action which may have selective significance, such as the fact that blood 
group A substance inhibits the growth of the influenza virus or that 
blood group A substance enhances the virulence of the typhoid bacillus 
for mice. As an example of another type of " second order " interaction, 
cholera vibrios are extremely susceptible to acid conditions, and one of 
the major mechanisms by which the body is protected against cholera is 
the inability of the cholera vibrios to survive the acid condition in the 
stomach. Thus, the evidence for differences between the ABO blood 
groups in the amount of stomach acid (K0ster, Sindrup, and Seele, 1955 ; 
Sievers, 1959) may indicate different susceptibilities to cholera. And 
finally there are many significant associations between blood groups and 
various diseases, which include filariasis (Franks, 1946), polio (Jung- 


blut, Karowe, and Braham, 1947), diphtheria (Kosling, 1928), scarlet 
fever (Nowak, 1933 ; Kubaschkin and Leisermann, 1929), measles (Miro- 
nesco, 1929), typhoid (Schapiro, 1932; Gorecki and Kulesza, 1949), 
whooping cough (Mironesco, 1929), and tuberculosis (Weinberger, 1943). 
For all of these diseases the evidence is certainly far from clear, but in 
a review of these investigations, I have tested the results by chi-square 
and have found many to be significant. 

To close this brief review of the evidence for selection on the blood 
groups, I would like to point out another recent discovery which I think 
may have selective implications. Nelson (1953) has demonstrated that 
the red blood cell plays an active role in the body's defense against disease. 
In the presence of complement and immune antibody, many bacteria, 
including those of pneumonia, diphtheria, dysentery, tuberculosis, strep- 
tococci, and also the treponema of syphilis, adhere to the surface of the 
red blood cell from which they are then phagocytized by the leukocytes. 
In the absence of the red blood cells the amount of phagocytosis is greatly 
reduced. Since the adherence of the bacteria to the wall of the red cell 
is not simply adsorption, it may well be that the surface antigens of the 
red cell would have some effect on this process. Hence different surface 
antigens may influence the ability of the red cell to perform this function, 
and if this is the case, then it may be the physiological mechanism by 
which the blood groups influence the individual's reaction to disease. 

Although the preceding review indicates some associations between 
the blood groups and disease, all this work is obviously not conclusive 
but circumstantial ; and circumstantial evidence never convinced anyone. 
But I think that the great mass of evidence at least indicates that this 
is an area that should be explored more fully. However, the great number 
of associations between the blood groups and diseases has even contri- 
buted to the skepticism about them, since in general one would not 
expect to find so many correlations. On the other hand, it seems to me 
that this great number of correlations may explain some of the pecu- 
liarities of the distribution of the ABO blood groups. Because of the 
great number of diseases involved, there would be a relative constancy 
of the frequencies over large areas, but since many of these diseases are 
epidemic in nature, there will be local variation when some communities 
are affected while others are not. In addition, in areas such as Europe 
where there has been much intercommunication and thus exchange of 
parasites, the frequencies should be more similar than in isolated areas. 
Much of the variation or the " aberrant " blood group frequencies which 


have been found in primitive populations are perhaps due as much to 
the diseases which ravaged them at the time of first European contacts 
as to random gene drift to which they have usually been ascribed. At 
any rate, although we are still far from a solution to the problems of 
the distribution of the ABO blood groups, I think recent developments 
bode well for the future. 


Boyd, W. C. 1939 Blood groups. Tabulae Biologicae, 17: 113-240. 

Brues, A. M. 1954 Selection and polymorphism in the A-B-0 blood groups. 

Am. J. Phys. Anthrop., 12: 559-597. 
Cavalli-Sforza, L. L. 1958 Some data on the genetic stucture of human popu- 
lations. Proc. of the Tenth Internat. Cong, of Genetics, Vol. 1, 389-407. 
Franks, M. B. 1946 Blood agglutinins in filariasis. Proc. Soc. Exp. Biol, and 

Med., 62:17. 
Gorecki, J., and S. Kulesza 1949 Typhoid fever and blood groups. Polish 

Med. Week, If. 1254-1257. 
Hirszfeld, L., H. Hirszfeld, and H. Brokman 1924 On the susceptibility to 

diphtheria with reference to the inheritance of the blood groups. J. 

Immunol., 9: 571-591. 
Johannsen, E. W. 1927 A classification of cancer patients according to their 

blood groups and some investigations concerning isohemagglutination. 

Acta Path, et Microbiol. Scandinavia, 4- 175-197. 
Jungblut, C. W., H. E. Karowe, and S. B. Braham 1957 Further observations 

on bloodgrouping in poliomyelitis. Ann. Intern. Med., 26: 67-75. 
Kimura, M. 1956 Rules for testing stability of a selective polymorphism. Proc. 

Nat. Acad. Sci., 42: 336-340. 
Koppisch, E., and J. Oliver-Gonzalez 1959 Agglutination of erythrocytes in 

vivo in mice after injection of ascaris extracts related to immunization 

with human erythrocytes. Proc. Soc. Exper. Biol, and Med., 100 : 827-829. 
K0ster, K. H., E. Sindrup, and V. Seele 1955 ABO blood groups and gastric 

acidity. Lancet, 2: 52-55. 
Mironesco, T. 1929 Les groupes sanguins dans les maladies infectieuses. Arch. 

d'Anat. Microscop., 25: 166-172. 
Nelson, R. A. 1953 The immune-adherence phenomenon. Science, 118: 733-737. 
Nowak, H. 1933 tiber Blutgruppen und die Empfanglichkeit fur Diphtherie und 

Scharlach. Zeitschr. Rassenphysiol., 6: 136-157. 
Parr, L. W. 1930 On isohemagglutination, the hemolytic index, and hetero- 

hemagglutination. J. Infect. Dis., 1^6: 173-185. 
Rosling, E. 1928 Undersogelser over Difteridisposition og Difteriimmunitet 

med saerlight Henblick paa Blodgrupper, Schickreaktion og Arvelighedds- 

forhold. Arnold Busch, Copenhagen. 


Rubaschkin, W. J., and L. I. Leisermann 1929 Blutgruppen und Krankheiten. 

Ukrainisches Zehtralblatt fur Blutgruppenforschung, 3: 303-311. 
Schapiro, A. 1932 Hematologische und serologische Zusammenhang mit den 

Blutgruppen. Zeitschr. Rassenphysiol., 5: 49-52. 
Sievers, M. L. 1959 Hereditary aspects of gastric secretory function. Am. J. 

Med., 27: 246-255. 
Snyder, L. H. 1926 Studies in human inheritance. The linkage relations of 

the blood groups. Zeitschr. Immunitat., J$: 464-480. 
Struthers, D. 1951 ABO groups of infants and children dying in the west of 

Scotland (1949-1951). Brit. J. Soc. Med., 5: 223-228. 
Weinberger, M. 1943 An investigation on blood groups in tuberculosis. Brit. 

J. Tubercul., 37: 68-87. 
Wright, S. 1951 The genetical structure of populations. Ann. Eugen., 15: 323- 




Department of Medicine, University of Washington 

Seattle, Washington 

THE study of human evolution until recent years was limited to 
investigations of the gross structural characteristics of man and 
dealt primarily with man's evolution from hominoid precursors. Such 
studies of long-term evolutionary adaptations have resulted in our present 
concept of man's ancestry. The human biologist in search of short-term 
evolutionary trends in man will need to look elsewhere for a field of inves- 
tigation. Genetically, human evolution may be conceived as occurring 
through changes in the frequency of genes in the human gene pool. 
As conceived today, the primary function of a gene is the determination 
of specific protein structure; all other gene effects appear to be conse- 
quences of this primary effect. It would therefore be profitable to study 
the gene frequency of genetically controlled proteins and protein enzymes 
since such traits would represent reasonably close primary gene products. 
Differential distribution of polymorphic traits 3 in different populations 
who have lived under varying environments should give some infor- 
mation about the possible adaptive values of such traits. There is 
increasing evidence that in most cases the existence of polymorphisms 
where the rarest trait exceeds a frequency of one or two per cent or more 
in a population implies that the trait represents the end result of some 
selective advantage for the trait carriers. Trait injurious to their 
carriers lower fitness, lead to a smaller number of offspring, and therefore 
will be eliminated by natural selection. Fisher (1930) has pointed out 

1 Investigations reported in this paper were supported by grants of the National 
Institute of Health, Commonwealth Fund and the Rockefeller Foundation. 

2 John and Mary R. Markle Scholar in Medical Science. 

3 A polymorphism in man is defined as the simultaneous existence in the same 
population of two or more distinct forms of a given characteristic in such propor- 
tions that the rarest cannot be maintained by mutation alone. A metabolic poly- 
morphism in man may affect the various gene-controlled proteins and enzyme 
variants such as hemoglobins, transferrins, haptoglobins, enzyme deficiencies, etc. 


that mutations to neutral traits without advantage or disadvantage would 
have a very small chance to become established in a population. Occa- 
sionally, however, chance or drift might lead to survival of neutral 

Natural selection, by causing differential survival of some genotypes 
at the expense of others, is recognized to be the most significant evolu- 
tionary agent. Natural selection in man has two broad aspects: 
differential fertility and differential mortality. Certain genes may lead 
to differential reproduction rates among different members of a popu- 
lation. Genes affecting fertility may interfere primarily with normal 
sperm, egg, or embryonic development, and thus lead to genetic death. 
Other genes, by affecting the emotional and social aspects of behavior, 
may interfere with reproduction in some individuals and not in others. 
It is not generally recognized that in western societies one-fifth to one- 
sixth of individuals of a given generation produce one-half of the next 
generation (Neel, 1958). In present western civilizations the genetic 
determinants of differential fertility are hard to untangle from cultural 
factors that lead to fertility reduction. Neel (1958) wisely pointed out 
the necessity of comparing fertility rates in civilized and primitive 
societies to elucidate the genetic component of differential fertility in 

Emphasis will be placed in this discussion on differential mortality. 
An infant mortality of 50% or more is a common phenomenon in 
primitive societies. Obviously, if genes existed which protected against 
dying, they would have evolutionary survival value. Such genes would 
tend to increase in a population and, unless exerting some other unfavor- 
able effect, would tend to replace their alleles. If, however, individuals 
die at random regardless of genotype, no significant change in gene 
frequency would be expected. What are the environmental features 
leading to high childhood mortality ? Infectious diseases (" pestilence ") 
and starvation have long been known to be principal factors checking 
population growth (Malthus, 1798). 


Is there evidence that starvation-induced mortality is selective? 
Knowledge of the genetic control of human nutritional requirements is 
not yet far advanced. However, judging from mammalian experiments, 
it is fairly certain that individuals differ genetically in nutritional require- 
ments for a variety of nutrients such as amino acids and vitamins 


(Williams, 1956; Schneider, 1958). There is also increasing evidence for 
the genetic control of obesity. Some individuals will be obese with 
identical caloric intake and energy expenditure while others will not 
(Tepperman, 1958). Since genes fundamentally control all enzyme- 
mediated reactions of intermediate metabolism, it is likely that death 
by starvation would not act at random, but certain genotypes would be 
more susceptible to famine-induced death than others. A human popu- 
lation that has lived through many famines, as have most, is therefore 
likely to harbor genotypes which are relatively resistant to death by 

A recently discovered polymorphism in man may conceivably be of 
pertinence. About three per cent of the Canadian and British population 
have a genetically determined relative deficiency of the serum enzyme, 
pseudocholinesterase (Kalow and Staron, 1957). This enzyme deficiency 
was discovered by noting that some patients developed prolonged apnea 
following administration of a drug, succinyldicholine, which requires the 
enzyme for breakdown. The fundamental lesion leading to pseudo- 
cholinesterase deficiency presumably is an altered enzyme protein mole- 
cule that no longer exerts its full enzymatic action. The primary 
function of pseudocholinesterase is not known. Its action on drugs 
such as succinyldicholine is most likely incidental to its unknown 
principal function. Assays for the enzyme utilize its property to over- 
come the effect of certain inhibitors. It has recently been observed that 
extract of potato peelings is a potent inhibitor of the abnormal enzyme 
(Harris and Whittaker, 1959). Since the potato is a relative newcomer 
to Western European foods (350 years), it is unlikely that the potato 
itself has played a role in producing this polymorphism. However, there 
may be other foods containing the unidentified ingredient of the potato 
peel which may have conditioned this polymorphism in the past. 

Infectious diseases. 

Haldane (1949, 1957) pointed out that infectious diseases probably 
have been the main agent of natural selection of man during the past five 
thousand years. As soon as the invention of agriculture and urbaniza- 
tion made relatively dense populations possible, such selective agents as 
vertebrate predation ceased to be important and diseases, spread by over- 
crowding, took their place as agents of natural selection. When man was 
first exposed to the typical human infectious diseases, capacities acquired 
by earlier natural selection, such as the nimbleness of mind and agility 


of body necessary to outfight predators and escape from them, failed to 
protect him against infectious-disease mortality (Allison, 1960a). How- 
ever, man's genotype usually was sufficiently heterogeneous that genes 
existed which protected their carriers against death from such diseases. 
With differential mortality of susceptible genotypes, consecutive epidemics 
of various infectious diseases acted as a sieve, concentrating resistant 
genes in those populations that were exposed. With a few exceptions 
(see below), the definite nature and action of such genes has remained 

Genetically determined resistance to mortality from infectious disease 
usually is not generalized but is highly specific for various diseases. 
Depending upon the preference for a given tissue and the mechanism of 
multiplication of the invaders, specific genes, acting on specific metabolic 
sequences and in different tissues, will protect the host against a given 
disease. Genetic resistance against disease produced by a given micro- 
organism usually does not protect against other micro-organisms. If 
micro-organisms are related, some protection usually carries over. An 
example of such a mechanism is the inherited virus-multiplication de- 
pressing factor in mice (Sabin, 1954) . This single gene-controlled factor 
prevents infection by a group of related viruses (yellow fever, W. Nile 
fever, Japanese B encephalitis, St. Louis encephalitis, Eussian spring 
tumor encephalitis). The identical gene, however, does not prevent 
infection by a different group of central nervous system viruses (Western 
and Eastern encephalomyelitis, poliomyelitis, rabies, lymphocytic menin- 
gitis, herpes virus, Rift Valley fever). All data (Gowen, 1948) indicate 
that the development of genetic resistance is a highly specific phenomenon 
acquired through evolutionary adaptation differing with the "medical 
history" of the population. 

One would expect widespread diseases with high mortality to have 
had the most pronounced effect as selective agents. Age at the time of 
death from infectious disease also is important since only diseases 
which kill before reproduction will exert significant evolutionary effects. 
Degenerative diseases such as arteriosclerosis and cancer, which kill after 
reproduction has occurred, cannot be selective agents unless otherwise 
associated with fertility differentials. 

Genetic factors causing resistance to infectious diseases should be 
carefully distinguished from acquired immunity. Immunity can be pas- 
sive and is caused by direct transfer of circulating antibodies to the baby's 
circulation from the mother. Such passive immunity will disappear a 


few months after birth. Active immunity implies the development of 
antibodies after contact with the disease agent. In contrast, genetically 
determined natural resistance involves the inheritance of gene-controlled 
characters which protect the individual entirely or partially against the 
disease or its effects, regardless of whether or not prior contact with the 
disease agent has occurred. Although the capacity to develop immunity 
also has some genetic determinants, the present discussion excludes 
acquired immunity and is concerned only with inherited factors of 
natural resistance. 

Evolution in host and parasite: myxomatosis as an evolutionary model. 

Carefully controlled studies in many mammalian species leave no 
doubt of the role of genetic factors in resistance to infectious disease. 
The literature in this field is considerable and has been summarized 
(Gowen, 1948 and 1951; Hutt, 1958). 

The study of infectious diseases and their impact on the evolution of 
man is made more difficult by evolutionary adaptations in both host and 
parasite. At the same time as genetically determined host resistance 
develops, the parasite tends to become less virulent by mutation. The 
end result may be a highly attenuated micro-organism living in a genet- 
ically resistant host. Could it be that man's orphan viruses or viruses 
to which no specific disease can be assigned are examples of such evolu- 
tionary adaptations ? Since the duration of a human generation is long 
as compared with that of man's parasites, evolutionary changes in micro- 
organisms producing human disease are probably frequently important 
in leading to less severe disease manifestation in a population. The 
main emphasis in this discussion, however, will deal with the changing 
genotype of the host. 

Eecent field studies on a viral disease of rabbits, myxomatosis, illus- 
trate admirably the development of genetically determined host and para- 
site resistance and may be considered as a model of how genetic resistance 
to infectious disease develops (Fenner, 1959; Marshall and Fenner, 1958). 
The myxomatosis virus when first introduced into the Australian rabbit 
population was highly virulent and killed more than 95% of the infected 
animals. It was noticed that mortality decreased from epidemic to 
epidemic. When the offspring of animals trapped in successive epidemics 
was infected with a uniform dose of a standard virus preparation, the 
percentage of animals killed decreased from year to year. There was an 
inverse relationship between mortality and the number of previous epi- 



demies the animals' ancestors had suffered. These data indicated the 
development of a raised frequency of heritable resistance factors to myxo- 
matosis. The heritability of these factors was demonstrated by breeding 
experiments. The exact nature of the genes producing genetic resistance 
to rabbit myxomatosis is not known. Active or passive immunity as the 
explanation of disease resistance to myxomatosis was carefully ruled out 
by the conditions of the experiments. 

Fio. 1, Interaction of Host Constitution and Bacteeial Constitution in 
Infectious Disease Mortality (redrawn after Gowen). 

Evolutionary changes in the virus were demonstrated by infecting 
highly susceptible standard laboratory rabbits with different myxomatosis 
virus preparations isolated in successive epidemics. In these experiments 
using a genetically stable host, a decreasing mortality from year to year 
also could be demonstrated, indicating the emergence of less virulent 
virus strains with successive epidemics. 


The population dynamics in these epidemics involves differential mor- 
tality which kills susceptible genotypes and leaves the more resistant 
individuals to contribute more offspring. After several years, a large 
proportion of rabbits is therefore genetically resistant. Genetic changes 
of the virus probably result from the selective advantage enjoyed by those 
virus mutants which do not kill the host and therefore do not die with the 
host. Figure 1, redrawn from an article by Gowen (1952), illustrates 
the interaction of host susceptibility and parasite virulence and the effect 
on mortality, resulting from the interplay of both factors. 

Infectious disease and balanced polymorphisms. 

In many instances genes which protect against an infectious disease 
may have lowered the fitness of the individual in other ways. This 
situation where a gene has both advantageous and injurious properties 
leads to "balanced polymorphism" where a fixed population frequency 
of the trait is reached (Ford, 1945; Sheppard, 1958; Allison, 1959a). 
This frequency represents a balance between the tendency for gene- 
frequency increase due to beneficial factors and for gene-frequency 
decrease from deleterious effects. It will be shown that three common 
red blood cell polymorphisms of man, sickling trait, thalassemia, and 
glucose-6-phosphate dehydrogenase deficiency, probably owe their present 
frequencies to selection by falciparum malaria. 

It may be argued on general genetic principles and on the example 
of the polymorphisms conferring resistance to malaria, that most genes 
protecting against infectious diseases initially were rare and deleterious 
(Allison, 1960a; Sheppard, 1958). Furtherfore, many presently known 
deleterious recessive and other genes, such as those for cystic fibrosis, 
phenylketonuria, spastic diplegia, schizophrenia, and hyperuricemia, exist 
in populations at such frequencies that chance or mutation pressure could 
not have accounted for their presently existing incidence (Penrose, 1957). 
They, therefore, must be considered true polymorphisms. Carriers of 
such traits presumably have enjo}^ed some advantage in the past. Since 
infectious diseases probably represent an important selective sieve in 
man's recent evolutionary history, it is not unlikely that many of these 
deleterious recessive genes may have been spread through populations by 
the selective action of epidemics. With disappearance of infectious dis- 
ease in recent years, traits with deleterious effects (balanced poly- 
morphisms), apart from their protection against death from infectious 
disease, will decrease in frequency once their selective advantage is 


removed. If most genes conferring resistance to infectious disease do 
indeed lower fitness, control of infectious diseases will lead to decreased 
frequency of many deleterious genes by removing their selective advan- 
tage. In medical terms, the result would be the decline of some genes 
that cause hereditary diseases. This is an important and somewhat para- 
doxical phenomenon demonstrating how environmental manipulation by 
improved hygienic conditions can lead to eugenic improvement. 4 

Those genes which confer resistance against infectious diseases but 
which otherwise have neutral effects, would be expected to have replaced 
their alleles through selective action. If the infectious disease disappears 
before such a trait has spread through the entire population, a given 
attained frequency would continue throughout future generations in 
contrast to the declining gene frequency observed with balanced poly- 
morphisms. A gene that protects against infectious disease and that has 
additional beneficial effects would spread through a population very 
rapidly and would even continue to do so at lesser speed when the 
infectious disease is no longer present. 


When tuberculosis first strikes a population without previous contact 
with the disease, mortality is high. In 1890 tuberculosis was first intro- 
duced into the Qu'Appelle Valley Indian Eeservation in Saskatchewan. 
The annual tuberculosis death rate reached the all-time high figure of 
almost 10% of the total population (Ferguson, 1955). More than one- 
half of the Indian families were eliminated in the first three generations 
of the epidemic. Twenty per cent of the mortality of the remaining 
population was also due to tuberculosis. After 40 years and three genera- 
tions most susceptible individuals apparently died and the annual death 
rate had been reduced to 0.2%. 

High mortality rates and acute disease had always been noticed when 
various populations were first exposed to tuberculosis (table 1). On 
genetic principles, exposure over many generations should lead to death 
of susceptible individuals with survival of those individuals who are 
genetically resistant. 

The Ashkenazi Jewish population of America represents a population 

4 It needs to be pointed out, however, that in rapidly expanding populations as 
exist today (doubling every 30 to 40 years) the actual number of patients with 
these diseases will actually increase, even though the percentage frequency of the 
genes determining these diseases will diminish. 


that has survived the high tuberculosis attack rate of the crowded ghettos 
of Europe for many generations. Although the rate of tuberculosis 
infection, as determined by tuberculin testing, has been identical in Jews 
and Gentiles, tuberculosis mortality is significantly less in Jews than in 
Gentiles (Perla and Marmorston, 1941 ). 5 In contrast to the acute 



Fulminant infection and high mortality on 1st exposure in: 

American Indians 


Puerto Ricans 



East Asians 


rapidly progressing caseous type of tuberculosis with regional lymph 
node involvement and the high rate of tuberculosis meningitis during 
early exposure of a population to tuberculosis, populations with a history 
of contact over many generations develop a more chronic, fibrous tuber- 
culosis, as seen now in Europe and in America. Yemenite Jews immi- 
grated to Israel from an agricultural milieu with little tuberculosis. 
Tuberculosis among the Yemenites is of the rapidly progressing type with 
a high mortality (Dubos and Dubos, 1952). Statistics collected during 
World War II in the ghetto of Warsaw suggest that the genetically 
acquired relative resistance of Ashkenazi Jews to tuberculosis can be 
rapidly overcome in conditions of extreme crowding and starvation 
(Dubos and Dubos, 1952). 

The importance of the host factor is underlined by the different 
reactions of human populations with different ancestral tuberculosis 
histories when exposed to the disease in the United States at the present 
time. Most of the populations listed in table 1, such as Indians, Eskimos, 
and Puerto Kicans, still demonstrate a much more acute type of disease 
when they contract tuberculosis. 

Several family studies suggest hereditary susceptibility (Puffer, 1946). 
Since, in a contagious disease, environmental factors are difficult to 
separate from genetic determinants, these data are difficult to evaluate. 

5 Some of this difference, however, may be due to better medical care enjoyed 
by the Jewish group. 


All twin studies, however, have demonstrated increased concordance for 
the disease among monozygotic twins as compared with dizygotic twins 
(table 2). 


Tuberculosis in twins {von Verschuer, 1959) 




Diehl and 

von Verschuer 








Uehlinger and Ktinsch 








Kallmann and Reisner 








Vaccarezza and Dutrey 
































Harvald and Hauge 







Total 381 202 843 187 53.0 22.2 

In view of the suggestive nature of these human data, conclusive 
animal experiments on genetic resistance to tuberculosis are of special 
interest. By artificial selection several teams of investigators have 
succeeded in raising strains of rabbits which are either highly susceptible 
or resistant to tuberculosis (Lurie et al., 1951 ; Diehl, 1958). Eesistance 
is unrelated to acquired immunity. It appears that resistance is caused 
by the inherited ability of the animal to phagocytose the tuberculosis 
organism, thus preventing spread (Lurie et al., 1951). The nature and 
action of the genes involved is unknown. 

The combination of all evidence suggests strongly that the present 
relatively high resistance of Western populations to tuberculosis is genet- 
ically conditioned through natural selection during long contact with 
the disease. The decline in tuberculosis mortality had begun before 
discovery of the tuberculosis organism and before medical measures were 
taken and probably is partially due to selective mortality of susceptible 
population members. 


Plague has been one of the great killers of populations in past times. 



During the European epidemic in the 14th century, a minimum of 25% 
of the entire European population died. In certain areas the mortality 
was much higher, as shown in table 3. Data from South Africa suggest 
that Europeans, as descendants of populations who survived the most 
severe plague epidemics, are more resistant to pneumonic plague than 
Negroes, while Negroes are more resistant than Asiatic Indians, Chinese 









Kollath (1951) 

Liibeck (Germany) 








Smolensk (Russia) 




Avignon (France) 




Marseille (France) 



Venice (Italy) 



Great Britain 


Pollitzer (1954) 

Rats — no plague for 30 yrs 


Sokhey and Chitre ( 1937 ) 

Rats — severe plague in 

recent yrs. 



and Malayans, who are most susceptible (Mitchell, 1927). Older accounts 
also mention the greater susceptibility of the Negro. The genetic signifi- 
cance of these data is doubtful in view of the difference in living con- 
ditions between the various populations. 

Although definite genetic resistance factors have not been proven 
in man, studies on rats suggest strongly that plague bacilli kill the more 
susceptible genotypes. Wild rats, the carriers of the human plague 
bacillus, were captured from many cities in India where the recent 
experience with plague differed greatly from city to city. These rats 
were inoculated with a standard dose of plague bacillus. Mortality in 
rats was inversely proportional to plague exposure, varying from 91% 
for rats from cities with no plague for the previous 30 years to 10% for 
rats from cities with severe plague up to two years before capture of 
the rats (Sokhey and Chitre, 1937). The design of these experiments 
appears to exclude active immunization and suggests the acquisition of 
genetic resistance. 




There is good reason to believe that smallpox contributed to the rapid 
defeat of the Aztecs by Cortez and his conquistadores. The disease 
apparently was transmitted to the Indians in 1520 by a Negro in Cortez' 
army and spread very rapidly through the population. Early writers 
give a mortality figure of 3,500,000 for this epidemic (Kollath, 1951; 
Dubos, 1959). It appears that at least one-half of the Indian population 
died. A century or so later repeated outbreaks of smallpox decimated 
many American Indian populations in North America. Early in the 17th 
century the Massachusetts and Narragansett Indians were reduced from 
30,000 and 9,000, respectively, to a few hundred. In the 19th century the 
Mandan population fell from 1,600 to 31 and very high mortality was 








appro. 50% 

Dubos (1959) 


American Indians 

( New England ) 

over 90% 

Dubos (1959) 




Perla and 
Marmorston (1941) 


American Indians 
( Mandan, Assiniboin, 



Dubos (1959) 




Hirsch (1883) 


Marquesas Islanders 


Hirsch (1883) 

noted among the Assiniboins, the Crows, the Plains tribes, and the Black- 
feet (table 4). In fact, the spread of smallpox probably was one of the 
first examples of biological warfare. The European settlers realized the 
high susceptibility of the Indians to smallpox and purposely spread 
infected blankets (Dubos, 1959). 

In Hawaii, in 1853, there were over 9,000 cases of smallpox with 
6,000 deaths among a population of 70,000 (Hirsch, 1883). In 1707 
after a long period of freedom from smallpox an epidemic in Ireland 
killed 18,000 out of 50,000 inhabitants (Perla and Marmorston, 1941). 
Protective immunization prevents smallpox in Western countries. The 
disease has become endemic rather than epidemic in areas of the world 


such as India where vaccination is not generally practiced. Case mor- 
tality — not population mortality — still averages 25%. The role of genetic 
resistance in this disease is difficult to evaluate but presumably has played 
a role. 


Measles is a rather benign disease with negligible childhood mortality 
at this time in Western society. The measles virus produced severe dis- 
ease which killed large numbers when first introduced into virgin popu- 
lations without previous contact. Panum described the introduction of 
measles into the Faroe Islands in 1846 (Panum, 1940) ; three-quarters 
of the population became infected. Eecent epidemics of this type have 
also occurred in isolated communities; for example, Eskimos of the 
Canadian Arctic suffered a mortality of as high as 7% (Dubos, 1959). 
The Tupari Indians of the Brazilian forest were first discovered in 1949 
and numbered some 200 people. Six years later two-thirds of the group 
had died of measles introduced by rubber gatherers (Dubos, 1959). 
Measles had a very high mortality rate in various Pacific islands during 
the last century. A Hawaiian king and his queen died of measles in 1824 
during a visit to England. Their attending physicians, including Sir 
Henry Halford, president of the Eoyal College of Physicians, found it 
hard to believe that a disease "which even a delicate London girl might 
bear could be so destructive to robust denizens of the Pacific" (Dubos, 
1959). In 1848 every child born in Hawaii died of measles, pertussis 
or influenza. In 1874, 20,000 Fiji Islanders died of measles introduced 
from Sydney, Australia (Perla and Marmorston, 1941). Since adults 
suffer a markedly increased mortality whenever they contact measles, 
some of the above findings probably can be explained on that basis. 
It is possible however that modern populations with low measles mortality 
have acquired genetic resistance against death from measles. The possi- 
bility of diminished virulence of virus strains undoubtedly plays a role. 


Genetic factors appear to play a role in susceptibility to paralytic 
poliomyelitis. Evidence along different lines is available : a. There is 
significantly increased occurrence of paralytic poliomyelitis in certain 
families in different years and in different generations (Aycock, 1942; 
Addair and Snyder, 1942). These data have been interpreted as com- 
patible with single recessive gene inheritance (Addair and Snyder, 1942). 


Paralytic poliomyelitis in isolated communities (Sabin, 1951) 






per 100,000 

Chesterfield Inlet 

Arctic Eskimos 








5300 dead 
many more paralyzed 
















St. Helena 




Nicobar Island 








808 dead 
many more paralyzed 




Poliomyelitis in Hawaii, 1988-19If1 (Sabin, 1951) 






Part Hawaiian 










Sabin's mouse data implicating a single pair of recessive genes in causing 
resistance to certain nervous system viruses are also of pertinence in this 
regard (Sabin, 1954). b. Twin studies show a high concordance rate 
among monovular twins (35% concordance in monozygous twins versus 
6% in dizygous twins and 6% in other siblings) (Herndon and Jennings, 


1951). c. Attack rates of paralytic poliomyelitis are very high in isolated 
inbred communities (table 5). d. Attack rates of clinically recognizable 
poliomyelitis vary in different population groups living in the same neigh- 
borhoods and attending the same schools without segregation (table 6). 
Although the case mortality of poliomyelitis usually is not high and 
does not compare in magnitude with some of the previously cited diseases, 
selective factors may play a role under primitive conditions where paral- 
ysis may be a serious handicap to life. As in other diseases, mutability 
of the virus, rather than the host, is again a most important factor in 
causing differences between epidemics. 

Yellow Fever and Trypanosomiasis. 

It has been argued that genetic host factors have played a role in the 
natural history of yellow fever (Sabin, 1954). The highly endemic areas 
of Africa are for the most part inhabited by a population group that does 
not suffer from the severe clinical manifestations of the disease. 
Strangers coming to these areas usually die of a severe form of the disease. 
It is likely that prolonged exposure has killed off the susceptible geno- 
types. The occurrence among South American Indians of both mild and 
severe forms of yellow fever is in accord with the hypothesis that 
yellow fever was imported from Africa after the discovery of America 
and has not had sufficient time to kill off all susceptible individuals. 

When Trypanosomiasis is first introduced into an area where it did 
not exist, it may kill one-third to two-thirds of the exposed population. 
After some years it becomes a much milder disease (Dubos, 1959). An 
interesting hemoglobin polymorphism exists in cattle which may have 
some bearing on trypanosomal resistance. Hemoglobin B in cattle is 
absent from the Mututu and N'Dama breeds of Nigeria which are more 
susceptible to trypanosomiasis than Zebu cattle (Bangham and Blumberg, 


This disease has had a profound influence on human events and mor- 
tality for at least 2,000 years (Boyd, 1949). Malaria is very widespread 
and even now is said to kill two million children every year. On evolu- 
tionary grounds, such a disease should be an important selective agent. 
Polymorphisms affecting body tissues necessary for malarial growth which 
exist in populations subjected to malaria for many generations (sickle- 
cell trait, thalassemia, glucose-6-phosphate dehydrogenase deficiency) 



would therefore be suspect of owing their distribution to the selective 
action of malaria. 

Several varieties of malarial parasites produce human malaria. The 
most potent selective agent would be that variety of malaria associated 
with the highest mortality. Falciparum malaria is the most lethal type 
of malaria and is found most frequently in tropical and subtropical areas 
(figure 2). Yivax malaria is less likely to kill and is found more 

Fig. 2. Distribution of Falciparum Malaria (see Boyd, 1949). 

frequently in temperate zones. On evolutionary and parasitologic 
grounds, vivax malaria is the older species (Knowles et al., 1930; Boyd, 
1949; Bray, 1957). The almost complete natural resistance of the West 
African Negro to vivax malaria is therefore of interest (Boyd, 1949). 
This resistance is unrelated to any of the known polymorphisms discussed 

The host tissue in which malarial organisms primarily proliferate is 
red blood cells. The malarial parasite has many of the enzyme systems 
that exist in the red blood cell and depends on some of the enzymes and 
metabolites of the red cell for its normal metabolism (Trager, 1957; 
Geiman, 1951). The adaptation of the parasite to the red blood cell 
represents a finely balanced end result of evolutionary development. One 


would expect suboptimal growth of malarial parasites in red blood cells 
which deviate from the normal, since an abnormal red cell might be a 
less satisfactory host for malarial growth. Since the number of parasites 
in the red blood cells appears to be related to malarial mortality, patients 
with other than normal red cells would be expected to have a lower 
malaria mortality rate. 

The development of acquired immunity in children living in holoen- 
demic malarial areas must be understood to evaluate the effect of possible 
genetic resistance factors against malarial mortality. An infant in a 
holoendemic malarial area will be born with passive immunity acquired 
from the mother. Passive immunity disappears after the first one-half 
year of life. The child then becomes susceptible to malarial death be- 
tween six months and the second to third year. At that time active 
immunity to the parasite begins to play a significant role in preventing 
mortality. Although older children may still be heavily infected with 
parasites, clinical illness becomes progressively milder and more rare so 
that such individuals appear to live more or less in harmony with the 
parasite (Macdonald, 1957; McGregor, 1959). A study attempting to 
show the protective effect of a genetic trait on malarial mortality there- 
fore should concentrate on young children in the age group of six months 
to three years. 

Sickle-Cell Trait. 

This trait is due to the single dose of a mutant gene causing pro- 
duction of an abnormal hemoglobin molecule — the sickling hemoglobin. 
The sickling trait is widely spread throughout equatorial Africa and is 
also found in Greece, Turkey, and India (figure 3). Sickle-trait carriers 
have both normal and abnormal hemoglobin in all their red blood cells. 
For practical purposes the sickle-cell trait is not associated with disease. 
Mating of two sickle-trait carriers produces 25% offspring with the double 
dose of the sickling gene. Under primitive conditions of life, this con- 
dition — sickle-cell anemia — is usually lethal before reproduction. Fre- 
quencies of sickling trait as high as 40% occur in certain African 
populations. Since with every death of a child from sickle-cell anemia, 
two sickle genes are lost from the population, the explanation for high 
sickle-trait frequencies would demand an abnormally high mutation rate 
or reproductive overcompensation. Both possibilities have been ruled out 
by direct studies (Vandepitte et al., 1955; Allison, 1956). An alternate 
explanation would postulate a selective advantage for sickle-cell trait 



carriers, balancing the loss of sickling genes from deaths due to sickle- 
cell anemia. It is now generally conceded that carriers of the sickle-cell 
trait are less likely to die from falciparum malaria. Properly controlled 

Fig. 3. Distribution of Hemoglobin S and Hemoglobin C. 

Sickling and malaria {Vandepitte and Delaisse, 1957) 



Malarial children 

(> 1000 parasites/mm 3 ) 


X 2 = 18.9 



1180 286 

P < 0.01 



studies of young children have demonstrated protective action of the 
sickling trait (Vandepitte, 1959; Vandepitte and Delaisse, 1957; Allison, 
1957; Eaper, 1959). Table 7 shows representative data on the relation- 
ship of parasite counts to sickling trait in a highly malarial population, 
demonstrating the protective action of the sickling trait against malarial 


proliferation. Other less extensive data have shown definitely diminished 
malarial mortality of sickling-trait carriers (Lambotte-Legrand and 
Lambotte-Legrand, 1958). More such data are needed. It has been 
calculated that a mortality from malaria of 10%, if the deaths occur 
entirely among the nonsickling population, is sufficient to explain the 
persistence of the sickle-cell gene at the high observed frequency (Allison, 
1956, 1957). If some trait carriers also die from malaria, the figure 
must be increased proportionately. The sickling trait does not necessarily 
protect against infection by malaria. In many studies the proportion of 
parasite carriers was found to be identical in sicklers and the normal 
population. From the evolutionary point of view, differential mortality 
is the important phenomenon and, indeed, has been found. 

Control of malaria should lead to decline of the sickling gene by 
removing the selective advantage. Decrease of sickle-cell trait frequency 
in fact has been demonstrated in Negroes of the Dutch West Indies whose 
specific origin in Africa was known (Jonxis, 1959). The lowered fre- 
quency of the sickle-cell trait in American Negroes as compared with 
African Negroes is also suggestive. The literature on the relationship 
of sickling to malaria has been extensively reviewed (Vandepitte and 
Delaisse, 1957; Allison, 1957; Lehmann, 1959; Neel, 1956, 1957). 

Other Abnormal Hemoglobins. 

Apart from Hemoglobin S, many hemoglobin variants have been dis- 
covered in the last few years. Most of these are rare mutants ; those that 
can be definitely classified as polymorphisms are Hemoglobin C and 
Hemoglobin E. 

Because of the limited geographic distribution of Hemoglobin C to a 
relatively small area of West Africa (figure 3) it has been postulated that 
the mutation leading to Hemoglobin C is more recent and may have 
developed from that of Hemoglobin S (Mourant, 1954). Hemoglobin C 
trait is not associated with illness. Since homozygous Hemoglobin C 
disease is far less lethal than sickle-cell anemia, the selective advantage 
enjoyed by Hemoglobin C carriers would need to be less for genetic 
balance to occur. Data failing to demonstrate a protective action of 
Hemoglobin C towards malarial parasite density (Edington, 1959), 
therefore, may not be crucial. A selective advantage towards falciparum 
malaria of lesser magnitude than that balancing the sickling gene would 
be extremely difficult to demonstrate by present methods. Such proof 
may be possible only by careful collection of extensive mortality statistics 



of the different genotypes. Such practices do not yet prevail in the 
malarial regions of Africa. 

Hemoglobin E is found throughout Southeast Asia. Its incidence is 
about 10% in Siam, Burma, and North Malaya, and as high as 35% 
in Cambodia. All these are highly malarious countries. Hemoglobin E 
trait is innocuous and Hemoglobin E disease is relatively mild. No 
information on malarial protection is available. 

Fig. 4. Distribution of Thalassemia (see Chernoff, 1959). 


The thalassemia trait is a red cell abnormality common throughout 
the Mediterranean area, the Near East, Arabia, India, Southeast Asia, 
China, the Philippines, and Africa (figure 4). It is likely that the 
diagnostic term, thalassemia, includes several different genetic entities 
all producing a similar hematologic phenotype. A fairly well defined 
thalassemia-like condition is characterized by the presence of large 
amounts of fetal (E) hemoglobin in heterozygotes. Thalessemia in a 
yet unknown manner interferes with effective hemoglobin synthesis so 
that trait cells are deficient in hemoglobin. In some carriers the defect 
may be mild ; in others significant anemia may occur. The homozygous 
condition — thalassemia major — is a highly lethal disease. 


Considerations of the geographic distribution of thalassemia and the 
severe loss of thalassemia genes from thalassemia major suggested to 
Haldane the possibility of a selective advantage of the trait to malaria. 
Not many critical studies have been performed to test this hypothesis. 
Data collected in Sardinia suggest very strongly that thalassemia hetero- 
zygotes are indeed protected against malaria. Ceppellini and Carcassi 
studied two racially identical populations, one, living in a severely malar- 
ious area showed a high frequency (20%) of thalassemia, while the other 
population living in a non-malarial mountainous area, showed very little 
thalassemia (Ceppellini, 1959). 

Glucose-6-Phospliate Dehydrogenase Deficiency (Primaquine Sensitivity). 

This polymorphism (Beutler, 1959) was discovered several years ago 
when it was noted that 10% of American Negro soldiers developed severe 
blood destruction when given a new anti-malarial drug — primaquine. A 
series of brilliant investigations performed at the University of Chicago 
resulted in the demonstration that the defect causing susceptibility to 
drug-induced hemolysis was deficiency of a red cell enzyme — glucose-6- 
phosphate dehydrogenase (Carson et al., 1956). Abnormalities in gluta- 
thione metabolism of affected red cells were also described (Beutler,1959). 
The trait is inherited as a sex-linked character with intermediate domi- 
nance (Childs et al., 1958 ). 6 Investigations in Italy and Israel demon- 
strated that favism (hemolytic anemia from ingestion of raw fava beans 
or inhalation of products of flowering fava bean fields) only occurs in 
individuals with an identical enzyme deficiency (Szeinberg et al., 1958a; 
Sansone and Segni, 1957). Many other drugs have been demonstrated to 
cause hemolysis in enzyme deficient patients. In the absence of drug or 
fava bean exposure, the trait is not associated with clinical symptoms. 
However, hemolytic anemia induced by bacterial or viral infections 
appears to be not uncommon in enzyme deficient individuals (Marks, 

The development of a rapid screening test for the enzyme deficiency 
(Motulsky and Campbell, 1960) allowed us to test many different popu- 
lation groups in Seattle as well as during field trips to the Belgian Congo 
and Sardinia in spring, 1959, and to Alaska in the summer of 1959. 

6 Population surveys on males (who have only one X chromosome) will give 
the gene frequency of the trait. Carrier females ( heterozygotes ) usually have 
intermediate but may have normal or low enzyme levels. Frequency data col- 
lected on females may therefore be misleading for genetic purposes. Incidence 
data on males are always bimodal and clearcut. 



Countries in which favism or drug-induced hemolytic anemia were 
reported are listed in table 8 (Beutler, 1959; Motulsky and Campbell, 
1960; Sansone et ah, 1959). Our studies and those of others on the 


Countries and islands with favism or 8-amino-quinoline induced 
hemolytic anemia or both 









Southern Italy 














Jews only ) 


distribution of the gene in different populations are summarized in table 9 
(Besides the studies cited in the table see also Sansone et ah, 1959, and 
Vella, 1959a). The limitation of the trait to a wide belt of tropical 
Africa, the Mediterranean area, the Near East, Indian Southeast Asia and 
the Philippines — all malarial areas (figure 5) is noteworthy. The trait 


Glucose-6-phosphate dehydrogenase deficiency 



American Negroes 


Motulsky & Campbell ( 1960) 


Leopoldville Negroes 


Vandepitte & Motulsky 


( unpublished ) 


Stanleyville Negroes 


Motulsky, Dherte & Ninane 
( unpublished ) 


Bayaka (S. Congo) 


Motulsky & Vandepitte 


(average 20%) 



Bwaka(N. W.Congo) 


Motulsky ( unpublished ) 



Motulsky (unpublished) 



Motulsky (unpublished) 



Motulsky ( unpublished ) 



Ninane & Motulsky (unpublished) 

S. African Bantus 


Charlton & Bothwell ( 1959 ) 

" Bushmen 


Bothwell ( personal communication ) 



Gilles, Watson-Williams & Taylor 




Asiatic Indians 













Eskimos (Alaskan) 


American Indians 


Peruvian Indians 


Oyana Indians 


Carib Indians 


Vella ( 1959b) , Motulsky & Campbell 

Motulsky & Campbell ( 1960 ) 
Vella ( 1959 ) , Beutler et al. ( 1959 ) 
Motulsky & Campbell (1960) 
Blumberg, Campbell & Motulsky 

( unpublished ) 
Walker & Bowmann ( 1959 ) 

Motulsky & Campbell (1960) 
Motulsky & Campbell ( 1960) 
Best (1959) 

Keller et al. (in preparation) 
Keller et al. (in preparation) 

Fig. 5. Distribution of Glucose-6-Phosphate Dehydrogenase Deficiency. 

has been found in South American Indians (Oyana), who have lived in 
malarial areas for many generations (Keller et al., in preparation). It 
is absent in Eskimos and some American Indians (Motulsky and Camp- 
bell, I960), and Peruvian Highland Indians (Best, 1959). It is also 
absent in northern Europeans, Japanese, and American Caucasoids 
(Motulsky and Campbell, 1960). 


Israeli workers demonstrated (table 10), a trait frequency of as high 
as 60% among Kurdish Jews, 25% among the Baghdad and Persian 
Jews, and lower percentages among various groups of Sephardic Jews, 
such as those from North Africa. The trait was extremely rare in the 
Ashkenazi Jews, who have not lived in malarial environments for about 
two thousand years. A high frequency among the Kurdish Jews prob- 
ably reflects a high degree of inbreeding in a relative small isolate. 


Glucose-6-phosphate dehydrogenase deficiency in Jews 

Szeinberg et al., 1958b, and Sheba, personal communication) 

Kurdish Jews 


Persian and Iraqui Jews 


Turkish Jews 


Yemenite Jews 


N. African Jews 


Ashkenazi Jews 


Thalassemia also has a high frequency among the Kurdish Jews. No 
enzyme deficiency was found among the Falasha Jews of Ethiopia where 
there is no malaria (Sheba, C, personal communication). 

Our field studies in Africa (with Dr. J. Yandepitte) demonstrated a 
high frequency of the trait in many tribes from malarial areas, and a 
low frequency in tribes with no malaria in the past (table 9). In view 
of the protective effect of sickling on malaria, the relationship of sickling 
and enzyme deficiency was of some interest. There was good correlation 
between frequency of sickling and enzyme deficiency. If the sickling 
rate of the tribe was low, then enzyme deficiency was also low and high 
enzyme deficiency frequency was associated with high sickling rates 
(figure 6). The only exception proved to be the Ituri Forest pygmies 
with a sickling rate of 31% while an enzyme deficiency rate of 4% was 
found. Sickling and enzyme deficiency affected different population 
members. The incidence of combined enzyme and sickling was not higher 
than expected by random coincidence of the two traits in an individual. 
The African data suggest that a common selective agent — malaria — has 
favored carriers of both these independent mutations. 

Field data collected in Sardinia (with Dr. M. Siniscalco) demon- 
strated, as in thalassemia, extremely high rates of enzyme deficiency in 
some coastal areas which were highly malarious until recently and low 



rates in the non-malarious central mountain areas (figure 7). Sardinia 
was until recently the most malarial area of Italy with the highest 
malaria mortality. Dr. Siniscalco and his group are carrying out further 
studies in Sardinia and other areas of Italy. 

Fig. 6. 



| 30- 




^ 10- 


2 34 

# I0 



D 10 

20 30 40 % 


ing Rate 

1. S. African Bantu 

7. Bashi 


8. Bayaka 

3. Usumbura 

9. Leopoldville 

4. Bahutu 

10. Stanleyville 


1 1. Pygmies 

6. American negro 

12. Nigerians 

Enzyme Def 


Data from Greece indicate that the enzyme deficiency is not infre- 
quent (Zannos-Mariolea and Chiotakis, 1959). The trait is especially 
common in the Patras area of the Peloponnesus and in the Chalkidiki 
Peninsula. These areas are also principal foci for sickling and thalas- 
semia and were known in the past for their high malarial endemicity. 
Greece is the only country in the world where sickling, thalassemia and 
glucose-6 -phosphate dehydrogenase deficiency occur at significant fre- 



quencies in an identical population. Usually, either sickling, such as in 
Africa, or thalassemia, such as in the Mediterranean area, Near East, 
India and S. E. Asia, have been found in populations affected by glucose- 
6-phosphate dehydrogenase deficiency. 

Fig. 7. 

Glucose-6-Phosphate Dehydrogenase Deficiency in Sardinia. 
(Siniscalco and Motulsky, unpublished) 

There is evidence that the enzyme deficiency in Africans may be 
produced by a different gene, since the mean level of enzyme among 
enzyme deficient Negroes is significantly higher than among enzyme 
deficient southern Europeans (Marks and Gross, 1959a; Motulsky un- 
published). It is not unlikely, therefore, that there are at least two sex- 
linked mutations (alelles?) both producing glucose-6-phosphate dehydro- 
genase deficiency. It is also possible that the different mean enzyme 


levels represent a phenotypic difference in red cells which in the two 
broad population groups constitute different genetic environments. 

Glucose-6-phosphate dehydrogenase deficiency does not affect all red 
cells equally. The oldest red cells (red cell life span is 120 days) have 
a much more severe degree of enzyme deficiency than do the younger 
red cells (Marks and Gross, 1959a and b). Since vivax malaria more 
readily parasitizes young red cells, it is unlikely that enzyme deficiency 
protects against vivax malaria. Although the selective advantage of the 
enzyme deficiency probably relates to falciparum malaria, it is signifi- 
cant that a not uncommon malarial parasite — plasmodium malariae — 
occurring with the same approximate geographic distribution as Plas- 
modium falciparum (Knowles et al., 1930) — is said to parasitize pref- 
erentially the older red cells. In view of the marked degree of enzyme 
deficiency in old cells, more detailed studies in a population where 
Plasmodium malariae is common might be of interest. 

Malarial organisms require glutathione for in vitro growth (Trager, 
1941; McGhee and Trager, 1950) and about 50% of the red cell's 
glutathione contributes to the cysteine requirement of malarial organisms 
(Fulton and Grant, 1956). Enzyme deficient red cells have a diminished 
amount of glutathione which is easily depletable (Beutler, 1959). There 
is some evidence that malarial organisms use the oxidative pathway of 
carbohydrate metabolism (Geiman, 1951) which is defective due to the 
enzyme deficiency. An enzyme deficient red cell would, therefore, be 
less likely to support optimal growth of malarial organisms than a normal 
red cell. To test this hypothesis directly, a field study was done with 
Dr. Vandepitte (unpublished) among 600 male Bayaka children less than 
10 years of age in a holoendemic malarial area of the Southern Congo. 
Parasite counts of enzyme deficient and sickling children were compared 
with those of normal subjects. The results were inconclusive and failed 
to demonstrate a statistically significant effect of enzyme deficiency or 
sickling on parasite multiplication as measured by parasite density. 

Allison and Clyde very recently have been able to demonstrate lower 
parasite levels in enzyme deficient children of East Africa in a field 
study limited to children between four months and four years (Allison 
and Clyde, 1960). For technical reasons Dr. Vandepitte and myself 
were unable to get many children in this age group, so that most of the 
individuals studied by us were five to ten years old. The difference in 
these results can be explained by immunity which in the older children 
blurs the critical differences. Using our test, Allison (1960b) also found 
a low frequency (1.7 to 2.9%) of enzyme deficiency in non-malarial 


areas and a high frequency (15 to 28%) in malarial areas of East Africa. 
All evidence therefore, strongly suggests that enzyme deficiency protects 
against malarial mortality. 

If from the point of view of natural selection, enzyme deficiency were 
entirely neutral apart from its effect on malaria mortality, we would 
expect that the trait would have replaced its normal allele in some 
populations. Since the trait exists as a polymorphism in all populations 
studied so far, the protective effect against malarial mortality must be 
counterbalanced by an injurious effect. Although favism does not appear 
to exist in Africa, it is likely that the enzyme deficiency sometimes may 
be lethal by causing blood destruction during infection. Glucoses-phos- 
phate dehydrogenase deficiency, therefore, appears to be balanced in a 
population by resistance to malaria on the one hand, and by other 
infectious diseases (e.g., common viruses) or foods (fava beans) which 
lower fitness on the other. Since the degree of lowered fitness in glucose- 
6-phosphate dehydrogenase deficiency undoubtedly is significantly less 
than that in sickle-cell anemia, the selective advantage necessary for 
balance needs to be only slight. The smaller the disadvantage of enzyme 
deficient carriers, the less of a selective advantage is required to explain 
the population frequencies of the trait. As in Hemoglobin C, critical 
studies may therefore be difficult to assemble. The detailed population- 
genetic implications of balance in a sex-linked trait such as the enzyme 
deficiency provide interesting problems and are under further study. 

Evolutionary Adaptation in Malarial Parasites; Polygenes. 

Since malarial parasites are known to undergo evolutionary adapta- 
tions readily, the possibility of mutations leading to parasite strains 
adapted to optimal growth in genetically abnormal red cells needs to be 
considered. A strain of parasites better adapted to sickle cells or enzyme 
deficient cells might theoretically emerge, although it has not been 
identified. Allison (1957) cites the following reasons why the normal 
strain might survive at the expense of the mutant. Subjects with the 
trait are always in the minority and usually represent a small fraction 
of the population. Furthermore, there is evidence in the sickle-cell trait 
that the normal parasite strain forms gametocytes 7 readily so that the 
normal strain would have as good a chance of infecting mosquitos as the 
postulated mutant strain. 

The demonstration that three specific traits, each under control of a 

7 That form of the parasite's life cycle which is ingested by the mosquito vector. 


single gene, appear to protect against malaria suggests how different 
genes might interact to produce a polygenic system of disease resistance. 
In fact, most data on genes conferring resistance to infectious disease in 
animals suggest the operation of polygenes. Judging from the relative 
resistance of the West African to vivax malaria, it is likely that many 
other yet unidentified genes exist which protect against malaria. Present- 
day success in isolating specific genes of a possible polygenic complex is 
a hopeful development indicating that analysis of polygenic traits in 
man may sometimes be approached with direct rather than statistical 


In contrast to the rather extensive data on the polymorphisms which 
presumably owe their distribution to malarial mortality, no data could 
be given on the nature of genes conferring resistance to other diseases. 
The reason obviously is the limited present-day possibility to study in- 
fectious diseases, such as smallpox or plague. Another reason relates 
to the pathophysiology of infection. With present methods we detect 
metabolic polymorphisms of body tissues that are easily obtainable, such 
as blood. When the various components of blood play a minor role or 
none at all in the production or dissemination of a disease, blood poly- 
morphisms are unlikely to play a role in disease resistance. Present 
techniques do not allow direct examinations of polymorphisms which 
affect the internal tissues of many individuals in a population. Some- 
times, however, blood cells may carry vestigial enzyme systems whose 
principal function is exerted in other organs. Study of such systems may 
allow a more ready approach to some of these problems. 

Serum Protein Polymorphisms and Infectious Disease. 

Eecently discovered polymorphisms of serum proteins are the hapto- 
globins and transferrin system. Haptoglobins are hemoglobin binding 
alpha 2 globulins, controlled by a single pair of autosomal genes (Smithies 
and Walker, 1955). Three electrophoretically distinguishable main 
varieties of haptoglobins exist (2 homozygotes and 1 heterozygote). 
Genetic suppression of haptoglobins occurs as a polymorphic trait among 
African and American Negroes (Giblett, 1959). Hemoglobin binding 
differs in the three varieties (Nyman, 1959) and is absent in ahapto- 
globinemia. Since the amount of haptoglobin increases in a variety of 
acute and chronic diseases, the fundamental physiologic function of 


haptoglobins may not be related to hemoglobin binding. In view of this 
adaptive response to disease in the individual, one might speculate that 
the basis of haptoglobin polymorphism may represent differential adapta- 
tion to infectious disease in the past. 

Transferrins are beta globulins which bind plasma iron. Ten percent 
of American Negroes and Australian Aborigines carry an electrophoret- 
ically detectable transferrin variant (Type CD1) (Giblett et ah, 1959). 
Other, more rare transferrin variants have also been described in Cau- 
casoid and Negro populations (Giblett et al., 1959). The basis for trans- 
ferrin polymorphism is not apparent from considerations of iron meta- 
bolism since the different variants bind iron to the same extent (Turnbull 
and Giblett, 1960). The demonstration that transferrin is a potent 
inhibitor of bacterial and viral multiplication (Martin and Jandl, 1959) 
suggests that this polymorphism may owe its origin to the selective action 
of infectious diseases in the past. 


Infectious diseases in man often produce a huge mortality when first 
striking virgin populations. Extensive evidence suggests that, apart 
from, and in addition to, immunity, genes exist which confer inherited 
resistance to a given infectious disease. The history and epidemiology 
of some of the great pandemics of man is reviewed with special reference 
to genetic resistance factors. It is considered likely that infectious 
diseases were one of the most potent agents of human natural selection 
in the past. Starvation is briefly discussed as another powerful agent 
of natural selection. 

A variety of chemical and enzymatic variants (metabolic polymor- 
phisms) exists in human populations at frequencies which only could 
have been reached with a selective advantage in the past. It is suggested 
that the present incidence of some metabolic polymorphisms has been 
caused by infectious diseases as selective agents. 

Host tissues essential for parasite growth or defense against the 
invading micro-organism would be the most probable site of such poly- 
morphisms. For example, genetically defective red cells, by limiting 
proliferation, might protect the host against an invader requiring red 
cells for multiplication. The only widespread and lethal human infec- 
tious disease of the red cell is malaria. There is good evidence that red 
cell defects, such as the sickle-cell trait and probably the thalassemia 
trait, protect against death from falciparum malaria. Eecent personal 


investigations are reviewed in detail, suggeging that another common red 
cell variant in tropical and subtropical populations — glucose-6-phosphate 
dehydrogenase deficiency of the red cell — also protects against falci- 
parum malaria mortality. 

The genetic implications of disease resistance are discussed in reference 
to the disappearance of infectious disease now and in the future. It is 
shown that the frequency of genes for disease resistance that have other- 
wise injurious effects, will decline in the future. Improved hygienic 
conditions, therefore, will lead to reduction of some human genes which 
produce illness. This effect has been observed with the sickling gene. 
Genes for disease resistance which otherwise are neutral will remain in 
the population at the frequency reached when the selecting infectious 
disease disappeared. 


Addair, J., and L. H. Snyder 1942 Evidence for an autosomal recessive gene 
for susceptibility to paralytic poliomyelitis. J. Hered., 33: 307-309. 

Allison, A. C. 1956 The sickle cell and hemoglobin C genes in some African 
populations. Ann. Hum. Genetics, 21 : 67-89. 

1957 Malaria in carriers of the sickle cell trait and in newborn children. 

Exp. Parasit., 68: 418-447. 

1959 Metabolic polymorphism in mammals and their bearing on prob- 
lems of biochemical genetics. Am. Naturalist, 93: 5-16. 

1960a Natural selection in human populations. U. of Kansas Science 

Bull, (in press) 

1960b Glucose-6-phosphate dehydrogenase deficiency in red blood cells 

of East Africans. Nature ( in press ) . 

Allison, A. C, and D. F. Clyde 1960 Malaria in African children with deficient 

glucose-6-phosphate dehydrogenase in erythrocytes. Brit. Med. J. (in press) 
Aycock, W. L. 1942 Familial aggregation in poliomyelitis. Am. J. Med. Sci., 

203: 452-465. 
Bangham, A. D., and B. S. Blumberg 1958 Distribution of electrophoretically 

different haemoglobins among some cattle breeds of Europe and Africa. 

Nature, 181 : 1551-1552. 
Best, W. R. 1959 Absence of erythrocyte glucose-6-phosphate dehydrogenase 

deficiency in certain Peruvian Indians. J. Lab. Clin. Med., 5If. 791. 
Beutler, E. 1959 The hemolytic effect of primaquine and related compounds: 

a review. Blood, Ik'. 103-139. 
Beutler, E., M. K. Y. Yeh, and T. Necheles 1959 Incidence of the erythrocytic 

defect associated with drug sensitivity among Oriental subjects. Nature, 

183: 684-685. 
Boyd, M. F. (editor) 1949 Malariology. W. B. Saunders, Philadelphia, 2 vols. 
Bray, R. S. 1957 Studies on the Exo-erythrocytic Cycle in the Genus Plas- 
modium. London School of Hygiene and Tropical Medicine. Memoir No. 

12. H. K. Lewis & Co., London. 


Carson, P. E., C. L. Flanagan, C. E. Ickes, and A. S. Alvinq 1956 Enzymatic 
deficiency in primaquine-sensitive erythrocytes. Science, 121^: 484-485. 

Ceppellini, R. 1959 Blood groups and haematological data as a source of ethnic 
information. In Medical Biology and Etruscan Origins. Edited by G. E. 
W. Wolstenholme and C. M. O'Connor. Little, Brown and Company, Boston. 

Charlton, R. W., and T. H. Bothwell 1959 The incidence of glutathione insta- 
bility of the red blood cells in the South African Bantu. S. Afr. J. of Med. 
Sc, 24: 88-89. 

Chernoff, A. 1959 The distribution of the thalassemia gene. Blood, lJf. 899- 

Childs, B., W. Zinkham, E. A. Browne, R. L. Kimbro, and J. V. Torbert 1958 
A genetic study of a defect in glutathione metabolism of the erythrocyte. 
Bull. Johns Hopkins Hosp., 102: 21-37. 

Diehl, K. 1958 Gestaltungsfaktoren bei der Tuberkulose in besonderer Beriick- 
sichtigung der Individuality des befallenen Organismus. Handbuch der 
Tuberkulose, I: 519. Thieme, Stuttgart. 

Dubos, R. 1959 Mirage of Health. Harper & Brothers, New York. 

Dubos, Rene, and Jean Dubos 1952 The White Plague: Tuberculosis, Man and 
Society. Little, Brown and Company, Boston. 

Edington, G. M. 1959 Some observations on the abnormal haemoglobin dis- 
eases in Ghana. In Abnormal Haemoglobins. Edited by J. H. P. Jonxis 
and J. F. Delafresnaye. Blackwell Scientific Publications, Oxford. 

Fenner, F. 1959 Myxomatosis in Australian wild rabbits — evolutionary changes 
in an infectious disease. The Harvey Lectures 1957-1958. Academic Press 
Inc., New York. 

Ferguson, R. G. 1955 Studies in Tuberculosis. University of Toronto Press, 

Fisher, R. A. 1930 The Genetical Theory of Natural Selection. Oxford Uni- 
versity Press, Oxford. 

Ford, E. B. 1945 Polymorphism. Biol. Rev., 20: 73-88. 

Fulton, J. D., and J. P. Grant 1956 The sulphur requirements of the erythro- 
cytic form of Plasmodium knowlesi. Biochem. J., 63 : 274-282. 

Geiman, Q. M. 1951 The cultivation of malarial parasites. In Parasitic Infec- 
tions in Man. Edited by H. Most. Columbia University Press, New York. 

Giblett, E. R. 1959 Haptoglobin types in American Negroes. Nature, 183: 192. 

Giblett, E. R., C. G. Hickman, and O. Smithies 1959 Serum transferrins. 
Nature, 183: 1589-1590. 

Gilles, H. M., J. Watson-Williams, and B. G. Taylor 1960 Glucose-6-phos- 
phate dehydrogenase deficiency trait in Nigerians. Nature, 185: 287. 

Gowen, J. W. 1948 Inheritance of immunity in animals. Ann. Rev. Micro- 
biology, 2: 215-254. 

1951 Genetics and disease resistance. In Genetics in the 20th Century. 

Edited by L. C. Dunn. Macmillan, New York. 

■ — 1952 Humoral and cellular elements in natural and acquired resistance 

to typhoid. Am. J. Hum. Genet., k : 285-302. 

Haldane, J. B. S. 1949 Disease and Evolution. Supplem. La Ricerca Scientifica, 
19: 68-76. 


1957 Natural selection in man. Acta Genetica et Statistica Medica, 6\ 


Harris, H., and M. Whittaker 1959 Differential response of human serum 
cholinesterase types to an inhibitor in potato. Nature, 183: 1808. 

Herndon, C. N., and R. G. Jennings 1951 A twin-family study of susceptibility 
to poliomyelitis. Am. J. Hum. Genet., 3: 17-46. 

Hirsch, A. 1883 Handbook of Geographical and Historical Pathology. The 
New Sydenham Society, London. 

Hutt, F. B. 1958 Genetic Resistance to Disease in Domestic Animals. Corn- 
stock Publishing Associates, Ithaca, New York. 

Jonxis, J. H. P. 1959 The Frequency of Haemoglobin S and Haemoglobin C 
Carriers in Curacao and Surinam. In Abnormal Haemoglobins, edited by 
J. H. P. Jonxis and J. F. Delafresnaye. Blackwell Scientific Publications, 

Kalow, W., and N. Staron 1957 On distribution and inheritance of atypical 
forms of human serum cholinesterase, as indicated by dibucaine numbers. 
Canad. J. Biochem. Physiol., 35: 1305-1320. 

Keller, J., X. Pi-Sunyer, T. Platjt, and P. A. Marks. Incidence of erythrocyte 
glucose-6-phosphate dehydrogenase deficiency among different population 
groups in Surinam (in preparation). 

Knowles, R., R. S. White, and B. M. Das Gupta 1930 Indian Medical Re- 
search Memoirs. Memoir No. 18. Thacker, Spink & Co., Calcutta. 

Kollath, W. 1951 Die Epidemien in der Geschichte der Menschheit. Verlag der 
Grief, Wiesbaden. 

Lambotte-Legrand, J., and C. Lambotte-Legrand 1958 Notes complementaires 
sur la drepanocytose. I. Sicklemie et malaria. Ann. Soc. Beige Med. Trop., 

Lehmann, H. 1959 Variations in human haemoglobin synthesis and factors 
governing their inheritance. Brit. Med. Bull., 15: 40-46. 

Lurie, Max B., Peter Zappasodi, A. M. Dannenberg, Jr., and G. H. Weiss 
1951 On the mechanism of genetic resistance to tuberculosis and its mode 
of inheritance. Am. J. Human Genet., Jf. 302-314. 

Macdonald, G. 1957 The Epidemiology and Control of Malaria. Oxford Uni- 
versity Press, New York. 

Malthus, T. R. 1798 Population: The First Essay. University of Michigan 
Press, Ann Arbor, Michigan. 1959 reprint. 

Marks, P. 1959 Proceedings of first Macy conference on genetics. Princeton 
( in press ) . 

Marks, P., and R. Gross 1959a Erythrocyte glucose- 6-phosphate dehydro- 
genase deficiency: evidence of differences between Negroes and Caucasians 
with respect to this genetically determined trait. J. Clin. Inv., 38 : 2253- 

1959b Drug induced hemolytic anemias and congenital galactosemia: 

examples of genetically determined defects in erythrocyte metabolism. 
Bull. N. Y. Acad. Med., 35: 433-449. 

Marshall, I. D., and F. Fenner 1958 Studies in the epidemiology of infectious 
myxomatosis in rabbits. J. Hygiene, 56: 288-302. 


Martin, C. M., and J. H. Jandl 1959 Inhibition of virus multiplication by 
transferrin. J. Clin. Inv., 38: 1024. 

McGhee, R., and W. Trageb 1950 The cultivation of Plasmodium lophurae in 
vitro in chicken erythrocyte suspensions and the effects of some constituents 
of the culture medium upon its growth and multiplication. J. Parasitol., 
36: 123-127. 

McGregor, I. A. 1959 The hyperendemic malaria of the Gambia. In Current 
Medical Research. Med. Res. Council. Her Majesty's Stationery Office, 

Mitchell, J. A. 1927 Plague in South Africa; historical summary (up to 
June 1926). Publications of South African Inst. Med. Res., 3: 89. 

Motulsky, A. G., and J. M. Campbell 1960 Rapid detection of glucoses-phos- 
phate dehydrogenase deficiency in red cells. Distribution and frequency of 
the trait and its possible relationship to malaria. Blood (in press). 

Mourant, A. E. 1954 The Distribution of the Human Blood Groups. Blackwell 
Scientific Publications, Oxford. 

Neel, J. V. 1956 Genetics of human hemoglobin differences: problems and 
perspectives. Ann. Human Genet., 21 : 1-30. 

Neel, J. V. 1957 Human hemoglobin types: their epidemiologic implications. 
New England J. Med., 256: 161-171. 

Neel, J. V. 1958 The study of natural selection in primitive and civilized 
human populations. In Natural Selection in Man. Wayne State Univer- 
sity Press. Detroit, Michigan. Reprinted from Human Biol., 30 : 43-72 

Nyman, M. 1959 Serum haptoglobin, methodological and clinical studies. Suppl. 
39. The Scand. J. of Clin, and Lab. Inv. 

Panum, P. L. 1940 Observations made during the Epidemic of Measles on the 
Faroe Islands in the Year 1846. F. H. Newton, New York. 

Penrose, L. S. 1957 Mutation in Man. In Effect of Radiation on Human 
Heredity. World Health Organization, Geneva. 

Perla, David, and Jessie Marmorston 1941 Natural Resistance and Clinical 
Medicine. Little, Brown and Company, Boston. 

Pollitzer, R. 1954 Plague. World Health Organization, Geneva. 

Puffer, R. R. 1946 Familial Susceptibility to Tuberculosis. Harvard Univer- 
sity Press, Cambridge, Massachusetts. 

Raper, A. B. 1959 Further observations on sickling and malaria. Trans. Roy. 
Soc. Trop. Med. Hyg., 53: 110-117. 

Sabin, A. B. 1951 Paralytic consequences of poliomyelitis infection in different 
parts of the world and in different population groups. Am. J. Pub. Health, 
41: 1215-1230. 

1954 Genetic factors affecting susceptibility and resistance to virus 

diseases of the nervous system. In Genetics and the Inheritance of Inte- 
grated Neurological and Psychiatric Patterns. Edited by D. Hooker and 
C. C. Hare. Williams & Wilkins Co., Baltimore. 

Sansone, G., and G. Segni 1957 Sensitivity to fava beans. Letter to the 
Editor, Lancet, 273: 295. 


Sansone, G., A. M. Piga, and G. Segni 1959 II Favismo. Edizione Minerva 

Medica, Genova. 
Schneider, H. A. 1958 What has happened to nutrition? Perspectives in Biol. 

and Med., 1 : 278-292. 
Sheppard, P. M. 1958 Natural Selection and Heredity. Hutchinson & Co., 

London, England. 
Smithies, 0., and N. F. Walker 1955 Genetical control of some serum proteins 

in normal human serum. Nature, 176: 1265-1266. 
Sokhey, S. S., and R. B. G. D. Chitre 1937 L'immunite des rats sauvages de 

l'Inde vis-a-vis de la peste. Bull. Off. Int. Hyg. Publ., 29 : 2093-2096. 
Szeinberg, A., C. Sheba, and A. Adam 1958a Enzymatic abnormality in ery- 
throcytes of a population sensitive to vicia faba or haemolytic anaemia 

induced by drugs. Nature, 181 : 1256. 

1958b Selective occurrence of glutathione instability in red blood cor- 
puscles of the various Jewish tribes. Blood, IS: 1043-1053. 

Tepperman, J. 1958 Etiologic factors in obesity and leanness. Perspectives 

in Biol, and Med., 1 : 293-306. 
Trager, W. 1941 Studies on conditions affecting the survival in vitro of a 

malarial parasite {Plasmodium lophurae) . J. Exper. Med., 7%: 441-462. 

1957 The nutrition of an intracellular parasite; avian malaria. Acta 

Tropica, U : 289-301. 

Turnbull, A., and E. GiBLETT 1960 The binding and transport of iron by 

unusual transferrins. Clin. Res. Proc, 8: 133. 
Vandepitte, J. 1959 The incidence of haemoglobinoses in the Belgian Congo. 

In Abnormal Haemoglobins. Edited by J. H. P. Jonxis and J. F. Dela- 

fresnaye. Blackwell Scientific Publications, Oxford. 
Vandepitte, JVM., W. W. Zuelzer, J. V. Neel, and J. Colaert 1955 Evidence 

concerning the inadequacy of mutation as an explanation of the frequency 

of the sickle cell gene in the Belgian Congo. Blood, 10: 341-350. 
Vandepitte, J., and J. Delaisse 1957 Sicklemie et paludisme. Ann. Soc. Beige 

Med. Trop., 37: 703-735. 
Vella, F. 1959a Favism in Asia. Med. J. Australia, 46: 196-197. 

1959b Susceptibility to drug induced haemolysis in Singapore. Med. 

J. Malaya, 13: 1-11. 

Von Verschuer, O. F. 1959 Genetik des Menschen. Urban & Schwarzenberg, 
Berlin, Germany. 

Walker, D. G., and J. E. Bowman 1959 Glutathione stability of the erythro- 
cytes in Iranians. Nature, 184: 1325. 

Williams, R. J. 1956 Biochemical Individuality. John Wiley, New York. 

Zannos-Mariolea, L., and P. Chiotakis 1959 Favism in Greece. Paper read 
before the European Congress of Hematology, London (typescript). 


University of Arizona, Tucson, Arizona 

MANY students of human evolution have made much of the lack oi 
specialized anatomical characteristics distinctive of our species. 
While it is true that too much emphasis may have been placed on this 
alleged lack of specialization (for the human foot is a uniquely specialized 
adaptation, and the human pelvic arrangements may certainly be so 
characterized as well), the fact remains that we are an extraordinarily 
versatile type of creature. Such specializations as we have appear to 
promote, or at the very least to be consistent with, versatile behavior, 
rather than to channel our abilities into a narrow field of supreme com- 
petence. A genetic organization which can produce phenotypes capable 
of such a wide range of activities as those performed by humans must 
have certain rather unusual qualities. 

It is the purpose of this paper to examine some aspects of the nature 
of such a genetic organization, in so far as it may be revealed by the 
phenotypes (human beings) which express it. It should be, I think, 
apparent upon a moment's reflection that the adaptation, or specialization 
if you will, having the greatest utility for a type of creature which must 
meet exceedingly varied circumstances cannot be one which would lead 
to the greatest degree of efficiency in mastering any one special circum- 
stance. In paradise this might be so, but we and the other living creatures 
of which we know live in the world of nature, where we must do the 
best we can with what we have, under circumstances which, sooner or 
later, are bound to change. Creatures which cannot fend for themselves 
under changed circumstances become extinct, as the whole fossil record 

Now it is a reasonably obvious fact that, in so far as humans are con- 
cerned, circumstances usually change sooner than later, and we aren't 
extinct yet. The variety of environments to which people must adjust, 
in one way or another, is enormous — certainly far greater than in the 
case of any other creature. Man's world-wide geographic range as well 
as the existence of diverse cultures make this so. These are just the 


conditions under which one might expect that of all characteristics, plas- 
ticity in bodily form would be adaptive. We might anticipate that the 
pressure of selection would work steadily in this direction, even though, 
as yet, we may be unable to define in just what manner such selective 
processes work. 

Perhaps, however, before getting deeper into the argument to be 
presented, it would be best to define the terminology which is being used. 
Adaptation is to be understood simply as the possession, or acquisition 
by a population, of a genetic system resulting in the existence of pheno- 
typic expressions which are favorable to survival and reproduction. The 
creatures which are members of such a population will not necessarily 
all possess the required phenotypes, but the population as such may be 
said to be adapted if : one, it demonstrates its ability to survive in com- 
petition with existing rivals for the same ecological niche; and if, two, 
it demonstrates its ability to survive the other hazards to its existence. 
These are clearly two separate problems which all populations have to 

For instance, a human population may succeed in surviving the 
stresses of disease, climate and the like in a given area under conditions 
of isolation, even though its adaptation to the local environment may be 
far from perfect, because the ecological niche which it occupies is so 
distinctive. If, however, a new human population should be introduced, 
which for any reason is better adapted to the local circumstances, it may 
drive the aboriginal group out of business. The fossil record indicates 
that such a sequence of events as this has frequently taken place among 
animals other than man. One aspect of adaptation, then, is a compara- 
tive one. 

Selection is a tricky term, since it may legitimately be used while 
discussing the results of competition between different populations, and 
also such competition as may take place within a population. Many 
authorities, furthermore, have attempted to make distinctions between 
natural selection, social selection, sexual selection, artificial selection, 
chance selection or genetic drift, and so on. For the purpose of this 
paper, the distinctions between types of selection are, in my opinion, 
irrelevant, although the concepts which the terms embody are important 
in other contexts. In the present context, selection may be understood 
simply as the survival and reproduction of certain alleles at the expense 
of contrasting ones, for any cause whatsoever. If populations differ from 
one another with respect to the incidence of such alleles, one whole 


population may be favored rather than the other. If, within a popu- 
lation, contrasting alleles exist, the same sequence of events can well 
occur, and a shift in gene-frequencies take place. 

Plasticity, finally, will be the term used to signify the possession of 
genetic systems on the part of the organisms concerned, which are capable 
of varied expression in response to varied environmental stimuli. In 
contrast to the adaptation, which is properly used only when speaking 
of populations ; and of selection, which may be used in discussing either 
populations or individuals, the term plasticity is appropriate only in 
connection with a discussion of individual organisms within a given 
population. One does not say that a population is plastic, but that it is 
variable, or polytypic, or polymorphic. 

Now adaptation to some particular set of circumstances is the end 
result of selection. Whole populations may suffer from this process and 
become extinct. Or gene-frequencies within a population may shift, so 
that it evolves, adapting to new circumstances as they arise. Either proc- 
ess results in the wastage of innumerable organisms, and the latter, given 
the mutation rate which appears to have existed in pre-Hiroshima days, is 
bound to be slow, so slow that evolution has rarely been seen in action. The 
length of time required for adaptation by selection to take place among 
human populations, may, however, be estimated by means of comparisons 
between the characteristics of different breeds of men who are known to 
have inhabited separated but similar parts of the world for a long time 
past. To be valid, such comparisons must be made holding the environ- 
mental conditions constant for all the groups studied, and using length 
of time in residence as the variable factor. Fortunately there are a few 
cases in which our knowledge of prehistory, dim though it may be, is 
still full enough to permit reasonable estimates not only of the length 
of residence of a given human population in an area, but also conclusions 
just as reasonable concerning the extent to which culture was able to act 
as a modifier of local environmental stresses. 

We have every reason to suppose, for instance, that human beings 
have inhabited such tropical parts of the Old World as Equatorial 
Africa, Southern India, and the islands off the Southeast tip of Asia, 
since the earliest palaeolithic — indeed, since a time when we would 
scarcely recognize them as human beings at all. During most of this 
period, the degree to which culture has provided adequate protection 
against jungle infections, solar radiations and the other tropical hazards 
has been minimal, and probably no greater in any of these regions than 


in the other two. Only during the last few hundreds of generations 
have drastic technological advances begun to be made in any part of the 
world, and for an even briefer time have they alleviated the conditions 
under which human beings in these areas had always been living. The 
native populations of Equatorial Africa, to a somewhat lesser extent those 
of Southern India, and people who in recent millenia have been limited 
to the farthest islands beyond Southwest Asia share so many biological 
similarities that they are commonly grouped as belonging to a hypo- 
thetical Major Eacial Stock called Negroid. As I have pointed out in a 
previous paper (Hulse, 1955) and as noted subsequently by Haldane 
(1956), the prior existence of any such stock as a discrete, unified entity 
is doubtful. On the contrary, the ancestors of the modern populations 
classified as Negroid have all been subjected to similar selective forces, 
in a tropical environment not much modified by cultural devices, for 
hundreds of thousands of years. That they should show similar adapta- 
tions in anatomy and physiology is to be expected. 

Equatorial America, in contrast, which presents the same sorts of 
hazards to human existence, has, by the best archaeological evidence, 
been inhabited by mankind only during the last twenty thousand years, 
more or less — certainly less than 5% of the time during which the deep 
tropics of the Old World have had a human population. Furthermore, 
the accepted theory is that the ancestors of its indigenes arrived after a 
long trek through the Arctic, and perhaps only after some degree of 
dependence upon complex technological aids to survival had come in 
existence. Despite the similarity of environment, the native populations 
of this part of the New World possess few if any of the characteristics 
which are used to classify people as Negroids. 

On the contrary, they share with other American Indians a large 
number of characteristics which have been described by some authorities 
(Coon, Garn, and Birdsell, 1950) as Arctic adaptations. This can be 
explained by supposing that selection-pressure requires, at a minimum, 
a great deal longer than twenty thousand years to produce adaptations 
suitable for tropical life, and at least a good many thousand years to 
eliminate the often sharply contrasting adaptations suitable to Arctic life. 

This conclusion is further substantiated by a review of the charac- 
teristics typical of the populations of the islands stretching out from the 
tip of southeast Asia. We have quite positive evidence of a continual 
flow, from the gene-pools of the more northern parts of the Orient, of 
wave after wave of populations absorbing and dispossessing the earlier 


inhabitants, doubtless due to the more efficient technology of the new- 
comers. Only such areas as Australia and Melanesia have not, as yet, 
been submerged in this flow of populations from the north. In Indonesia, 
as in the Amazon basin, the existing peoples are not Negroid. Native 
Australians, however, like Melanesians, often have been so labeled. The 
length of time during which the ancestors of the Australian aborigines 
have lived in temperate rather than tropical environments has been 
inadequate, despite their meager and primitive technology, to have as yet 
produced adaptations more suitable to their present homeland, except, 
possibly, for their skin-temperature regulating mechanism. 

Yet study of some bodily characteristics shows that far more rapid 
alteration of the phenotypes has taken place. Individuals of European, 
as well as of Asian and aboriginal American ancestry are likely to weigh 
a good deal less if they have grown up in hot places than if they have 
grown up in places where there is an annual period of cold. Sometimes 
they are slighter in build. Often they are shorter in stature. The 
difference in weight, however, appears to be that which is most fre- 
quently found; United States servicemen from the northern part of the 
country weigh, on the average, six pounds more than those from the south 
(Newman, 1956). The average weight of women of French ancestry in 
Vera Cruz (Montemayor, 1956) is 120 pounds, whereas that of French 
women in Paris (Felice, 1958) is five pounds more. The mean weight 
of Chinese in Canton is 18 pounds less than that of Chinese in Peking 
(Eoberts, 1953). Eskimos have a mean weight 20 pounds greater than do 
Maya Indians (Eoberts, 1953). One can multiply such examples almost 
endlessly. It would appear that gross body weight is quite responsive 
to differences in climate. The regression of weight upon mean tempera- 
ture is much less (Y60 to 66) among Mongoloids than it is among 
Europeans and Africans (Y85 to 88), however, as is shown clearly by 
Eoberts (1953). The importance of the length of time spent in the 
present habitat seems to be revealed by this fact. 

Weight is also, of course, responsive to other elements of environment 
as well, and these sometimes produce an effect contrary to that to 
be expected from temperature. Italian-Swiss in California weigh, on the 
average, 176 pounds; their close relatives in Switzerland weigh 16 pounds 
less (Hulse, 1958), although Swiss winters are far more severe. This 
cannot be regarded as a climatic adjustment. The fact is that in this 
carefully examined case, measured by the author, the migrants and 
American-born agree in weight closely, and both differ markedly from 
those examined in the homeland. The rapid response to the environ- 


ment found in this case is just a most extreme case of bodily plasticity. 
Evolution has not taken place. There is no evidence of a shift in gene 
frequencies in the populations concerned. Instead, we are forced to the 
conclusion that humans of most, if not all, stocks inherit the ability 
to alter their weight to a considerable degree in accordance with varied 
exterior stimuli. 

In other words, it is the quality of plasticity, not the weight, which 
is transmitted genetically. The extent of such plasticity, and the upper 
and lower limits of its range undoubtedly differ in different populations. 
Fat pygmies don't weigh as *^uch as skinny giants. But in all groups 
which have been observed, the amount of plasticity observed is very 
considerable, nor can it all be attributed to the ability of the rich to fill 
their bellies more readily than their poor cousins. Within any popu- 
lation, individual weights are quite varied. Some individuals weigh too 
much for their own good, and others weigh too little. But weights which 
fall within the range which permits effective action in meeting the 
problems and hazards of the individual in his own life are what we can 
expect to find and what, it appears, we do find much more often than not. 

It may seem that all this has been dwelling on the obvious. But it 
is important to reaffirm the differences between the type of variability 
which depends upon a genetically-based polymorphism, and the type of 
variability which depends upon a genetically-based plasticity. Since 
we find so many examples of each, in our study of the human body and 
its operation, confusing one with the other is all too easy, and this has 
often been done to the detriment of our understanding of human biology. 
Coon (1954), for instance, in his discussion of climate and race, uses 
such examples as I have given above to demonstrate the relationship 
which he believes to exist between these two. It seems to me that, in 
doing so, he simply side-steps the matter of race, a concept which must, 
to have any scientific utility, be based upon genotype rather than pheno- 
type. The fact that we must depend upon a study of phenotypes to 
learn about genotypes is irrelevant, and no excuse for a cavalier dismissal 
of the ancestral factor in a study of human variation. 

It has been shown, for instance, that Mongoloids in the tropics, 
whether in Borneo or Brazil, share the characteristics of light weight to 
some degree with other tropical peoples. The body build most prevalent 
among Mongoloids elsewhere is, however, to a considerable degree main- 
tained. Although not as thick-set as North Chinese or Sioux, perhaps, 
they do not approach the gracile types so often found among Miotics, 
Tamils, or Carpentarians. Rather, lessened weight among Mongoloids in 


the tropics appears to be an aspect of their shorter stature. Any photo- 
graph which shows Negroes in South America and South American 
Indians together shows the different ways in which the plasticity in 
bodily form may operate to reduce weight. 

I believe that we may well suspect that, although plasticity, in such 
a case as this, operates to help the adjustment of the individual to his 
environment, it is not as effective an adjustment in the long run, for 
the population of which he is a member, as adaptation by means of 
selection. Only prolonged, well devised and carefully conducted tests 
of the relative constitutional efficiency of Negroes and American Indians 
living in the same region could give an answer to this problem, of course. 
Be that as it may, plasticity may logically be expected to inhibit and 
delay the selective pressure which would otherwise lead to adaptation. 
It must be remembered that, until a few generations ago, the Indians 
of tropical America had, as competitors, only other Indians just as plastic 
but no better adapted than themselves. With the introduction of a popu- 
lation of African origin, the situation changed, and the success of 
populations of Negro origin in adjusting to the circumstances of life in 
the tropical parts of America cannot be denied. To what degree this 
is due to cultural rather than constitutional factors, I do not know ; but 
it is hard to believe that a background of slavery would have been of 
any assistance to the Negroes who escaped, or whose ancestors had 
escaped, from that condition. 

It would be interesting, it seems to me, to examine populations 
known to be of a mixed origin, part Indian and part Negro, in the 
malarial regions of tropical America. In this way it might be possible 
to discover whether genetic characteristics, such as the sickle-cell trait, 
which it has been alleged on good evidence have a positive selective value 
in such areas, have reached a higher frequency than have such charac- 
teristics as woolly hair, gray hair, baldness, blood types B, Henshaw, N, 
and Eh° in which Negroes and Indians are known to be widely different. 
Should this prove to be the case, it would not only add very strong support 
to the thesis that the high frequency of sickle-cell trait in some parts of 
Africa is indeed an adaptation, but it might also help to explain the 
degree of biological success which Negro and Negroid populations have 
attained in some parts of tropical America. 

This leads us to a further point, which needs to be made before 
proceeding any further. Although plasticity is certainly one of the out- 
standing characteristics of the human constitution, there are numerous 
characteristics in which it does not exist at all. Studies, by many scholars 


during the past fifty years, in which migrants from one environment to 
another are compared either to their offspring, or to their sedentary 
kinfolk at home, or to both, have been perhaps our best source of 
information concerning the extent of bodily plasticity among humans. 
They have, uniformly, demonstrated that, in such characteristics as may 
be called dimensional, and especially longitudinal, spending one's growth 
period in the United States results in larger bodily measurements. 
Studies by Boas (1911 et seq.) on various immigrants in New York, 
Shapiro (1939) of Japanese in Hawaii, Lasker of Chinese (1946) in 
eastern United States cities and Mexicans (1952, 1954) in various 
states, Goldstein (1943) of Mexicans in Texas, Hulse (unpublished), 
Suski (1933) and Greulich (1958) of Japanese in the Pacific Coast 
Kegion, Hulse (1957) of Italian Swiss in California and many others, 
are all in complete agreement in this respect. In bodily stature, and in 
measurements which may be regarded as fractions of stature, individuals 
raised in the United States are bigger than are related individuals raised 
elsewhere. Plasticity is very obvious. 

On the other hand, a study of the distribution of ABO and Eh 
blood-types among the American-born Japanese of Seattle shows the 
same frequencies as are found in the prefectures in Japan from which 
their parents came. A similar survey, also in Seattle, of persons with 
Irish surnames revealed the same frequencies as those of Irish in Dublin. 
Australia and New Zealand, populated almost entirely by people of 
British stock, have the same distribution of blood-types as is found in 
Britain, according to the tables published in Boyd (1939), Mourant 
(1954), and elsewhere. Migration into a new land, which has so greatly 
affected stature, has not led to any alteration in this respect. It is con- 
ceivable that selection, over the generations, may lead to a shift in gene 
frequencies among these populations. But this would be adaptation, 
not plasticity. 

At the same time it should be noted that in other traits in which 
genetic polymorphism is apparent, such as eye color, hair texture, and 
the like, we again find no difference between Italian Swiss of Californian 
birth, migrants to California, and sedents in Switzerland (Hulse, 1958). 
The changed environment has in no way altered these characteristics. 
Material for a really exhaustive analysis, world-wide in scope, is not in 
hand, but it would appear that characters of which the mode of inherit- 
ance is relatively simple, such as eye color, blood type, albinism, abnormal 
hemoglobins and so forth show no plasticity under changing environ- 
ments. On the contrary, characters of which the mode of inheritance 


is so complex that it has up to now defied analysis, do show some degree, 
at times a remarkable degree, of plastic response to changing environ- 
mental stimuli. Assuming no error in genetic diagnosis, simplicity in 
mode of inheritance is of course indicative of a rather direct genie action : 
a single gene leading to the development of a character; whereas com- 
plexity is equally indicative of several or many genes all playing 
important roles in the development of a character. 

Thus it would seem that, as a general hypothesis, we might say that 
the greater number of genes of which the joint action is required in 
order for a given phenotypic character to appear, the greater the chance 
that such a character will be plastic. The phenotypic expression, in 
such a case, might be expected to respond readily to changes in environ- 
mental opportunity and stress. The varying degree of expressivity of 
all the genes concerned would be to some extent effective in deter- 
mining the final result. A variety of aspects of the environment might 
be expected to impinge upon the line of development of the character. 
With the lines of genie action coming from many genes, more oppor- 
tunities for skewing or shifting the balance between them would logically 
seem to exist. Thus, we might expect not only that the chance of plas- 
ticity would be greater, but that the degree or extent of plasticity might 
also be greater, if several genes contribute to the phenotypic expression. 

Let us assume, for instance, that only one gene is concerned in the 
development of a character. Any environmental effect, say a difference 
in temperature, has only one point of attack upon the line of development 
which may lead to a modification of the phenotype. If, however, we 
assume that two genes are concerned, there will be two opportunities 
for environment to work on the development of the character; and 
whereas the first might be affected only by temperature, the second might 
be subject to some other type of environmental effect, perhaps an ample 
supply of oxygen. A shift in either environmental condition might then 
produce the same effect upon the phenotype, whereas a shift in both 
would result in a more drastic change. Should there be three genes at 
work, the number of different phenotypes to be expected under differing 
environments would be even greater. Continuing along the same line 
of thought, it would seem reasonable to suppose that the possibility and 
extent of plasticity would be still further enhanced if more genes than 
one, among those controlling the character, existed in more than one 
allelic form. Let us assume, for instance, that genes at four loci all 
contribute to the appearance of a character in the phenotype. If more 
than one allele exists at any of these loci, a still further degree of 


variability will certainly be introduced. In fact, it is easy to see how 
a continuous variation might be expected to be found, as it is, in dimen- 
sional characteristics, under such circumstances, in contrast to the dis- 
continuous variation which is apparent in such discrete characters as 
blood-types. There are, indeed, a number of characteristics not con- 
sidered as dimensional, such as skin pigmentation, which show a con- 
tinuous variation. Such characters are known, by now, to be the end 
result of the interactions of several genes, entirely aside from any environ- 
mental effect upon the phenotypes. Skin pigmentation also responds to 
alteration in the environment in most individuals. This has been shown 
to be the case among Negroes as well as other racial stocks (Lee and 
Lasker, 1959). Skin pigmentation itself serves as a prime example of a 
characteristic which, although in most living individuals it is somewhat 
plastic, in a number of populations would appear to be a genetic adapta- 
tion. Here too, confusion has arisen among taxonomists and students of 
the relationship between race and environment. Eelatively few indi- 
viduals lack the ability to tan; a larger number find this a slow and 
painful process. At the other extreme, we find whole populations among 
whom the deepest pigmentation is almost universal. Plausible if not 
necessarily accurate hypotheses have been advanced to explain the cases 
of deep pigmentation. It has been harder to explain the numbers, 
especially in populations of Northwest European ancestry, of individuals 
who just will not tan, although the need for vitamin D has been offered 
as reason for the existence of fair tannable skins among peoples living 
in a cloudy or foggy region (Coon, 1954). I would like to suggest that 
Allison's (1955, 1956) explanation of the high frequency of the sickle- 
cell condition among some African tribes may suggest a proper solution 
to this problem. 

If, indeed, fair but tannable skins have an adaptive value in cloudy 
environments, where synthesis of vitamin D by the skin would need to 
be maximal (Coon, 1954), and if, at the same time, a number of genes 
are concerned in the development and deposit of skin pigmentation, so 
that alleles leading to lack of such pigmentation may be found at several 
loci, then we may expect that, in order to supply the existing percentage 
of individuals with fair skins, gene frequencies for lack of pigmentation 
at each locus must be reasonably high. In random mating, a certain 
fraction of individuals must be procreated, who, like those Africans homo- 
zygous for sickle-cell anemia, really suffer. They combine too many genes 
for depigmentation and, when exposed to bright sunlight, are unable to 
tan ; but if unusually dark individuals have as great a disability to repro- 


duce, the genetic equilibrium will be maintained. To account for the 
difference in degree of pigmentation between Negroes and Whites, Stern 
(1953) came to the conclusion that genes at five or six loci were most 
likely. It may be that an equal number would be adequate to explain 
the differences within the European stock. If we assume five genes each 
with two alleles of equal frequency, and no dominance, but simply additive 
effect, it may be seen that .001 of the population would be utterly lacking 
the ability to tan, and an equal number very readily tannable or even 
permanently dark skinned. More than 99% would be more or less readily 
tannable with exposure to sunlight — in other words, plastic in their 
response to the environment. The greater the number of genes involved, 
the smaller the proportion of individuals who are involved in a real genetic 
misfortune. Selection pressure, then, should operate in favor of in- 
creasing the number of loci involved in this process. 

For a long time this quality of plasticity was overlooked by most 
scholars in physical anthropology. Since most of the human subjects 
who were examined and tested had grown up in the same places and 
under the same circumstances as their parents and grandparents, it was 
easy to assume that all resemblances between the generations were depen- 
dent upon genetic factors. It was only after Boas (1911) published his 
epoch-making work on the changes in bodily form among the descendants 
of immigrants that the extent of plasticity became at all apparent. It is, 
indeed, by the study of migrants, their sedent relatives, and their off- 
spring, that we can most effectively attack the problem of the degree and 
types of plasticity. Populations long resident in a given area may be 
genetically adapted, or plastically adjusted to their circumstances, or both. 
We cannot tell. Only when individuals shift from one environment to 
another is a variable introduced into the equation. Changing circum- 
stances in the homeland may result in changes in some bodily charac- 
teristics : the recent increase in stature, following improved sanitary and 
dietary conditions in many countries, is an example. But we can have 
far more precise genetic control in migrant studies. For this reason such 
studies appear to offer the best hope of disentangling the interacting 
factors of genetics, selection and environmental stress, and discovering 
in which characteristics plasticity appears to be greatest. 

Kaplan's (1954) paper on plasticity summarized the existing state 
of knowledge concerning changes in physical dimensions among migrants 
and their offspring so well that it would be fatuous to repeat her material 
here. There is no question but that human stature can increase, in a 
single generation, by as much as three per cent, under conditions which 


are favorable to growth. Such an increase in stature may be expected to 
be accompanied by increases in other dimensions, especially longitudinal 
ones, although the percentage increase is normally less in some and more 
in others. Proportions, too, may be altered: a slight decline in the 
cephalic index is noted in most such studies, for instance. It must be 
regretfully confessed that skin pigmentation has not been carefully 
enough recorded, in most cases, for us to calculate the degree, if any, of 
its alteration in a single generation. 

Whether or not these changes may all be regarded as examples of 
human plasticity, that is, phenotypic adjustments to the environment, 
is another question. Dahlberg, a generation ago, suggested that the 
greater size of the offspring of migrants might well be an example of 
heterosis. He supposed that migrants would be less likely to marry fellow 
villagers than would sedents, and that, consequently, their offspring 
should be heterozygous at a greater number of loci, having then, a greater 
growth potential. Fortunately the data which I collected on Italian 
Swiss at home and in California were suitable for a test of this hypoth- 
esis, and it appears to be true. About two-thirds of the actual migrants 
were the offspring of village endogamous matings — and from very small 
villages, averaging only a few hundred inhabitants each. Only about 
one-third of their California-born offspring were born of such matings, 
however. Among those who had remained in Switzerland the propor- 
tions were about even. In all three groups, the mean stature of men 
whose parents came from different villages exceeded by two centimeters, 
or more, the stature of men born of village-endogamous marriages, and 
their other longitudinal dimensions increased in proportion. A change 
in the social environment has made this increased growth possible, but 
it may not be regarded as a biological adjustment, only a biological result. 
Heterosis occurs in Switzerland as well as California, but more frequently 
in the latter because of changed social circumstances. Thus the mean 
stature of California-born would have exceeded that of their parents, even 
had there been no plastic response to the better living conditions of that 
state. That is, heterosis contributes a part of the greater size of the 
California-born. Hence, it is a factor which must be taken into account 
in migrant studies. 

Heterosis also suggests a possible solution to one of the outstanding 
puzzles in what is ordinarily called the " Eacial History " of Europe : the 
sudden increase in the incidence of brachycephaly, especially during the 
early Middle Ages. This increase is more than amply documented by 
numerous cranial collections, but it appears that a reverse trend is now 


going on (Sauter, 1957). It will be recalled that the offspring of migrants 
are not only taller, bnt usually more dolichocephalic than their parents. 
It also appears that village endogamy is associated with shorter stature — 
and consequently rounder heads, since stature is positively correlated 
with head length but not with head breadth. Now for a good many 
centuries previous to the Middle Ages there had been extensive movements 
of population in most parts of Europe during the period sometimes 
labeled the Volkerwanderung ; and during the period of the Eoman 
Empire retired legionaries were, more often than not, settled in colonies 
far from their birthplaces. Village or band endogamy would not be 
promoted by these circumstances. While the medieval structure of 
society lasted, however, serfs became attached to the land, and feudal 
lords looked with disfavor upon such divided loyalties as might be pro- 
duced by contact with the serfs of other feudal lords. Village or hamlet 
or manor endogamy should have been promoted by such a system, and I 
at least wonder whether a loss of stature due to an increasing frequency 
of consanguinous matings might not have been one of the biological 
results. There does, in any case, seem to be much evidence of reduced 
stature during this period, so that it is not surprising that we find, also, 
a higher percentage of brachycephalic individuals in the population. 

An alteration of head shape, if brought about by such a cause as this, 
cannot be regarded as an example of adaptation, plasticity, selection or 
evolution. It is simply a fortuitous result of a general depression in 
length measurements, which could have been brought about by any one 
of several causes. In the case of the offspring of migrants, the increase 
in stature is at least partly due to plasticity. There has been, so far as 
can be determined, no shift in gene-frequencies whatsoever. In the case 
of the Swiss, for instance, as mentioned above, the frequencies of alternate 
phenotypes in such items as eye color, hair color, hair form, and nasal 
profile remains the same among those of endogamous and exogamous 
parentage, just as the Japanese in Seattle retained the same gene-fre- 
quencies for blood-groups as their cousins at home. What we learn from 
a study of migrants and their offspring is that the overall size of the body 
is somewhat plastic, and the alterations of the size produce secondary 
results upon shape which may easily be mistaken for something else. 
Having learned this, we are now in a position to apply our knowledge to 
the analysis of other problems involving a study of the phenotype, such 
as racial history and the taxonomic relationship between tribes now 

We do not, however, dare overlook the fact that adaptation, despite its 
naturally slow rate, and despite the fact that plastic adjustments may 


inhibit it still further, must certainly still be taking place. Not all 
humans who are born live long enough to mate, and not all of those who 
do reproduce in equal quantity. Shifts in gene-frequencies would seem to 
be practically inevitable under such circumstances. Neel (1958) has 
pointed out the results to be expected in fertility itself, for instance. The 
hazards to our existence, and the problems of our lives, are certainly 
different from those of other animals and of our ancestors. Hazards and 
problems do remain, however, and afflict different individuals, and dif- 
ferent populations, to varying extents. Many of the shifts in gene 
frequencies which result may be fortuitous results of unique events, as, 
for instance, the unquestionably higher incidence of alleles for blondness 
in the 20th as compared to the 15th century in the population of the 
world at large (Hulse, 1955, 1957). Many others, however, may be 
presumed to have some adaptive value. The reduced incidence of sickle 
cell among peoples who, like Negroes in the United States, no longer 
suffer from malaria as their ancestors once did, would be an example 
of this (Allison, 1956). Perhaps the higher reproduction rate of women 
with small noses and of men with larger heads, found by Spuhler (1959) 
in Michigan, may reflect some adaptive value — at least in that part of 
the world. 

Eecently I had occasion to start investigating the later destiny of 
some 1200 Italian-Swiss who had been measured, 25 years previously, 
by Schlaginhaufen (1946) at the period of their first army service. Only 
about half of these men had fathered children, at least according to the 
vital statistics records, although about two-thirds of them had married. 
In this group I have, as yet, found no signs of any differential fertility 
in relation to stature, head-size or shape, nose-size or shape, hair-form 
and color, or eye-color. During the present generation, at least, genetic 
stability appears to exist despite the differential reproductive performance 
within the native population of the Canton Ticino. The plastic response 
to improving living conditions, and the increasing incidence of village 
exogamy, may be altering the phenotype to some degree. But no evolu- 
tionary trend can be either observed or predicted on the basis of the data 
on hand. 

Yet another disturbing factor has to be considered in any analysis 
of data derived from a study of migrants: the possibility of selective 
migration. Not everyone migrates, and we dare not assume in an off- 
hand way that migrants are genetically a perfectly random sample of 
the population from which they are derived, since we know that they 
are not a random sample either demographically or sociologically. As a 
matter of fact, most studies appear to show that migrants are slightly 


taller than sedents even if the migration has been only to another part 
of the same country. Martin (1949) found this to be true in Great 
Britain, Shapiro (1939) among the Japanese in Hawaii, and Lasker 
(1952) among Mexicans. As the last of these has so carefully docu- 
mented, however, the migrants usually left home before the completion 
of their growth period, so that selection need not be assumed. However, 
during my reexamination of Schlaginhaufer's data, I noted that those 
individuals who migrated after their period of army service, had already, 
before migrating, a mean stature 1.7 centimeters taller than those who 
were still living in their home canton 25 years later. The augmentation 
of growth in a new environment is apparently not the only factor to be 
taken into account, even though it is clear that it is the most important 
one, at least if one goes to the United States. 

Furthermore, we still are in total ignorance of the relative impor- 
tance of the different sorts of environmental stress which influence those 
characteristics in which man's body is plastic. Climate obviously plays 
a part. Nutrition obviously also plays a part. But no one has yet 
succeeded in measuring their relative importance. Indeed, American 
scholars have perhaps been a bit ethnocentric in their bland assumption 
that the diet of Swiss and Japanese in California, Mexicans in Detroit 
and Texas, and so on, must naturally be better than the diet of their 
sedentary cousins back home. Maybe it is, but I have not found figures 
to demonstrate it. I have found, however, that the Japanese who grew 
up in Hawaii 30 to 40 years ago were fed on as close an approximation 
to a Japanese diet as their mothers from the old country could devise. 
Quantification of the data proved impossible; it may be that they ate 
more than their cousins in Japan. Certainly the Italian Swiss in 
California fare better, especially with respect to meat and milk, than 
their cousins in Switzerland, but I am not yet in a position to measure 
precisely how much better. 

There seems every reason to believe that the state of health of an 
individual during the period of growth is also significant in the deter- 
mination of the size finally attained. But here again our understanding 
of the relative importance of plasticity and adaptation has been delayed 
by a total absence of the proper data. Complete medical histories for 
the period of growth of all the subjects examined in migration studies 
would be a minimum requirement for a proper approach to this problem. 
They are lacking. It would appear that childhood diseases tend to inhibit 
or at least delay growth, but we do not know whether such diseases are 
more prevalent among the stay-at-home cousins than among the offspring 
of migrants. The plastic response of the human body to conditions of 


health has been demonstrated. But we cannot demonstrate, as yet, that 
this fact is relevant to any differences between migrants or their children 
on the one hand, and se dents on the other. Nor, if it is relevant, do we 
hi*ve any idea of how important it may be. 

Clearly we are only at the beginning in our studies of the type and 
extent of bodily plasticity in the human species. That a certain degree 
of plasticity exists has been well demonstrated. That it affects most 
greatly those phenotypic characters dependent for their appearance upon 
the action of several or many genes appears to be a theory which best 
fits the facts as we know them. That, in general, this plasticity has an 
adaptive value of a high order for creatures which occupy our ecological 
niche, or zone, or plateau, seems likely. That a plastic, phenotypic 
adjustment should serve to slow down a shift of gene-frequenices in the 
direction of specific adaptations seems inevitable. That plasticity always 
leads to an improved adjustment between organism and environment, 
however, remains to be demonstrated. Presumably, for instance, the 
ability to put on weight, when possible, served a worthwhile function for 
our early ancestors who had no corner grocery store. To most of us 
nowadays, as we grow older, it appears to be an embarrassment, and is 
often, in fact, a contributory cause of death. 

Insofar as plasticity does lead to a better adjustment, however, it may 
be expected to render adaptation less necessary, and therefore to slow 
down the rate of evolution in the direction of genetic specialization. 
At the same time mutations leading to a greater degree of plasticity are 
likely to be selected; for, if our environment remains as changeable and 
diverse as it now is, this may be essential. One might anticipate then 
that plasticity would continue to operate in the future, as it has in the 
past, to maintain the genetic unity of the human species. 


Allison, A. C. 1955 Aspects of polymorphism in man. Cold Spring Harbor 
Symposium, Quant. Biol., 20 : 239-252. 

1956 Sickle-cells and evolution. Scientific American 195: 87-94. 

Boas, F. 1911 Changes in Bodily Form of Descendants of Immigrants. Govern- 
ment Printing Office, Washington, D. C. 

Boyd, W. C. 1939 Blood groups. Tabulae Biologicae, 17: 113-240. 

Coon, C. S. 1954 Climate and race in Climate Change (ed. H. Shapley). Har- 
vard University Press, Cambridge, Massachusetts, pp. 13-34. 

Coon, C. S., S. Garn, and J. Birdsell 1950 Races. C. C. Thomas, Springfield, 

Felice, Suzanne de 1958 Recherches sur l'anthropologie des Franchises. 
Masson et Cie., Paris. 


Goldstein, M. 1943 Demographic and Bodily Changes in Descendants of 

Mexican Immigrants. Inst, of Latin-American Studies, University of 

Texas, Austin. 
Gbeulich, W. W. 1957 A comparison of the physical growth and development 
of American-born and native Japanese children. Am. J. Phys. Anthrop., 

n.s. 15: 489-516. 
Haldane, J. B. S. 1956 The argument from animals to man: an examination 

of its validity for anthropology. J. Royal Anthrop. Inst., 86 (II) : 1-14. 
Hulse, F. S. 1955 Technolological advance and major racial stocks. Human 

Biol., 27: 184-192. 
1957 Some factors influencing the relative proportions of human racial 

stocks. Cold Spring Harbor Symposium, Quant. Biol., 22: 33-45. 

1958 Exogamie et heterosis. Archives Suisses d' Anthropologic Generate, 

22: 103-125. 

Kaplan, B. A. 1954 Environment and human plasticity. Amer. Anthrop., 56: 

Lasker, G. W. 1946 Migration and physical differentiation: a comparison of 

immigrants with American-born Chinese. Am. J. Phys. Anthrop., n. s. 4 : 


1952 Environmental growth factors and selective migration. Human 

Biol., 24: 262-289. 

1954 The question of physical selection of Mexican migrants to the 

U. S. A. Human Biol., 26: 52-57. 

Lee, Marjorie M. C, and G. W. Lasker 1959 The sun-tanning potential of 

human skin. Human Biol., 31 : 252-260. 
Martin, W. J. 1949 The physique of young adult males. Medical Research 

Council Memorandum No. 20, His Majesty's Stationery Office, London. 
Montemayor, F. 1956 La Poblacion de Vera Cruz. Gobierno de Vera Cruz. 
Mourante, A. E. 1954 The Distribution of the Human Blood-Groups. Black- 
well Scientific Publications, Oxford. 
Neel, J. V. 1958 The study of natural selection in primitive and civilized 

human populations. Human Biol., SO : 43-72. 
Newman, R. W. 1956 Skinfold measurements in young American males. 

Human Biol., 28: 154-164. 
Roberts, D. F. 1953 Body weight, race and climate. Am. J. Phys. Anthrop., 

n.s. 11: 533-558. 
Sauter, Marc R. 1957 Personal communication. 
Schlaginhaufen, 0. 1946 Anthropologica Helvetica. Arch. I. Julius Klaus 

Stiftung 21, Zurich. 
Shapiro, H. L. 1939 Migration and Environment. Oxford University Press, 

New York. 
Spuhler, J. N., and P. J. Clark 1959 Differential fertility in relation to body 

dimensions. Human Biol., SI: 121-138. 
Stern, C. 1953 Model estimates of the frequency of White and near-White 

segregants in the American Negro. Acta Genetica et Statistica Medica 


Suski, P. M. 1933 The body-build of American-born Japanese. Biometrika, 
25: 323-352. 




Department of Anatomy, Wayne State University 
College of Medicine, Detroit 7, Michigan 

MIGRATION and isolation play a significant role in ongoing human 
evolution. Both affect gene frequencies in various ways. Migra- 
tion brings to a geographical area genotypes evolved elsewhere; within 
a species it tends to increase intragroup genetic variability while reducing 
the intergroup differences in kinds and frequencies of genes. By 
subjecting individuals to changes in environment, migration also has the 
indirect effect on evolution of selecting for survival plastic genotypes that 
are capable of adequate response in a variety of environments. Over 
many generations and with repeated migrations it may lead to a sum- 
mation within the same individuals of characteristics adaptive under 
different conditions. 

Isolation, on the other hand, tends to lead to diversification of groups. 
Through local natural selection under local conditions it makes for the 
development of local subgroups. Isolation also permits local differentia- 
tion of groups on the basis of traits that have no adaptive significance. 
It even enhances the chances of survival of a group that may be handi- 
capped in comparison with other subgroups from which it has become 
genetically isolated. This may occur through the accidental selection of 
genotypes, a process which is called random genetic drift. 

Migration and isolation are the two faces of the same coin. Or 
perhaps it would be better to describe them as a coin and its mold, 
for isolation in its effect is the converse of migration. Thus the smaller 
the immigration rate, the greater the degree of isolation. Similarly, the 
product of the immigration rate multiplied by the population size is a 
coefficient of relative isolation (though technically one might prefer to 
reserve the name "index of isolation" for the reciprocal of this figure). 

In the present symposium our discussion is limited to biological 
evolution, more especially to change in frequency or kind of genes avail- 
able in a population. In this context the word "gene" is used in the 
broad sense and stands for any kind of transmitted genetic information. 


In respect to such effects, human migration is the genetically mean- 
ingful movement of people; that is, change of residence by individuals 
young enough to have offspring in the new location. In fact, the best 
way of measuring the extent of the effects of immigration is to count 
the proportions of individuals born and living to maturity in a certain 
place, both, one, or neither of whose parents were born there. From a 
genetic point of view the proportion of foreign to domestic gametes in a 
breeding population calculated from such data may be called the " effec- 
tive immigration rate." 

The scale on which migration is studied is of significance. It is one 
thing if the immigration to a large nation is 10% per generation and 
quite another if a similar rate of immigration is noted in a little farming 
village with only 100 adults in the population. An immigration rate of 
10% per generation to a major continental region may be viewed as a 
breakdown of racial distinctiveness and a trend toward genetic conver- 
gence of the population to that of other continents. 

In the small village, however, another force for differentiation will 
be at work, random genetic drift. Immigration of 10% may well be 
consistent with increasing intercommunity genetic differentiation. In a 
series of communities in a circumscribed region on the north coast of 
Peru, I studied several with total populations today of 3,000 to 15,000 
inhabitants. They were formerly smaller. Some degree of physical 
distinction persists in the people of these towns in the face of effective 
immigration rates of the order of 10 to 30% per generation. 

Thus in San Jose, a fishing village with an effective immigration rate 
of about 25% per generation during the last 50 years, one finds a 
relatively homogeneous population. The average individual is short in 
stature, has a large wide torso, big wide face, and long convex nose. Most 
of the people are Indian in appearance; and, despite the local history 
of continual coastal dwelling, they have the large thorax which is charac- 
teristic of Andean Indians. In Monsefu, a farming town with craft 
industries, the people are equally dark in hair, eyes, and complexion, but 
there is more variety in most measurements; and the average size is 
greater, the face smaller, and the nose more often short and straight or 
concave. In Mochumi, another farming town, the people are taller, 
more varied in appearance and more Mestizo, that is, more European in 
type, than those of either of the other two towns. Besides the differences 
in average measurements between the three towns, there are a number 
of differences between the offspring of native and non-native parents in 
each town. The offspring of immigrants to the towns are in general 


more variable. The process of increase in intracommunity variation and 
decrease in intercommunity differences thus continues to the present time 
but has not obliterated the traces of earlier population differences. 

In 1908, when Zangwill coined the phrase "the melting pot" to 
describe the flood of immigrants from diverse places and their assimila- 
tion and mixing in the United States, some 800,000 immigrants arrived. 
Nevertheless, for the United States as a whole, there has probably never 
been even a single generation with an immigration rate higher than that 
recorded in relatively isolated Peruvian farming and fishing districts. 
Thus immigration rates similar to those which were viewed as leading 
to assimilation or to the formation of enormous foreign enclaves in a 
nation can be seen as consistent with the maintenance of racial distinc- 
tiveness in a farming or fishing village. Of course, the situation is not 
comparable. In one small U. S. city the average pair of mates were born 
approximately 300 kilometers from each other, whereas in Mexican 
towns the average distance between the birthplaces of mates is 15 to 50 
kilometers (Lasker, 1954) and in Peruvian towns some 10 to 70 kilo- 
meters. The genetic effects of similar immigration rates must be similar 
if the original differences are comparable, but, in fact, the contrast of 
Spanish, Indian and Negro elements in North Peru is sharper than that 
between various European immigrants to North America in the early 
twentieth century. For the large population one may also tend to project 
trends over a long span of time. 

Thus, another axis along which migration may be examined is the 
degree to which the pattern of migration persists. A steady immigration 
or traditional exchange of mates between neighboring populations has 
a different significance from that of a unique event which may scatter 
Jewish displaced persons once a millennium or carry the Mongols to 
the gates of Vienna or the troops of Alexander to the Indus just once 
in recorded history. Of course, these great events and other historical 
movements have had their racial effects. But it is only the steady move- 
ments of people, manifest in general patterns, which can be described 
by abstract statistical formulations of the principles of genetic change 
and racial evolution. In the cataclysmic population movements which 
result from wars and natural calamities, the anthropologist can some- 
times find reasonable explanations of present population differences. He 
cannot hope to go far toward predicting future evolution in this way. 

In fact, the quantitative study of migration as a process in evolution 
going on from generation to generation in the world today is effectively 


limited to a small fraction of the kinds of effects which migration may 
have on changes in gene frequencies. 

Yet it provides valuable insights. Even the direct genetic effects 
of steady immigration pressure on small communities has been relatively 
little studied. In the past, when evolution was largely considered a study 
of biological history and the purpose was taxonomic, large-scale migra- 
tions and unique events served best to explain specific relationships. 
Mongoloid traits among European peoples were explained as the result 
of the settling in Sweden of Tatar troops, the invasion of Hungary by 
Asiatic Avars, and the raping of European women by hordes of Huns 
and masses of Mongols. 

In the discussion of migration so far, I have taken it for granted 
that populations of communities and countries differ in their gene fre- 
quencies and that a potential mate from one's own locality is more apt 
to share one's own genetic characteristics than is a potential mate from 
another place. The degree of isolation is thus one cause of assortive 
mating — matches in which particular characteristics of one mate are 
associated with particular characteristics in the other mate. The more 
isolated the community is, the more likely will husbands and wives share 
the same genetic traits. Men and women are most likely to marry 
persons in the same area or country; to this extent there is geographic 
isolation. Similarly, human beings usually marry members of their 
own social group : those who share the same values and speak the same 
language. Such social proximity may break down barriers of physical 
distance. For instance, after repeal of the Chinese Exclusion Law and 
passage of the so-called "Brides Act" of 1947, some 6,000 young Chinese 
Americans rushed to China to woo and marry before the right to bring 
Chinese-born wives to the United States expired in December, 1949. 
Most of the men went to the very counties where their ancestors had lived 
and sought introductions through the usual family channels; they mar- 
ried the very girls they might have married had their fathers never left 

Assortive mating also may be a deliberate or a subconscious choice of 
mates with similar characteristics. Folklore is full of instances which 
illustrate sexual attraction of physical difference — the beautiful dark-eyed 
Venus of Teutonic myth and the blond Hollywood siren of contemporary 
American society, with its predominance of brunettes. But for perma- 
nent bonds of matrimony most people choose a mate who resembles them- 
selves. Such choices narrow the number of potential mates and tend 
to limit the size of the breeding group ; hence they enhance the likelihood 


of maintaining within the group genetic distinctions from the population 
at large. 

Some human breeding groups are small enough for random genetic 
drift to be a significant factor. Virtually all the 200-odd inhabitants of 
the little island of Tristan da Cunha in the South Atlantic are descended 
from 8 men and 7 women who settled there after the Napoleonic wars 
(Kuczynski, 1953). The British colony of Pitcairn Island in the Pacific 
was peopled by the descendants of a few of the mutineers of the British 
ship " Bounty" and several Tahitian women (Shapiro, 1929). A study 
of a similar-sized group in the United States, one of the little religious 
sects, a Dunkard community in Pennsylvania, shows significant dif- 
ferences in gene frequencies from both the surrounding peoples in 
America and the peoples of the area in Germany whence their ancestors 
came (Glass et at., 1952). Nor are the gene frequencies intermediate 
between those of Americans and Germans. Furthermore, there is 
an apparent random genetic drift in three successive generations of 
Dunkards — studied synchronically, however (Glass, 1956). Other possible 
examples of random genetic drift — the systematic tendency to genetic 
homogeneity caused by the repeated possibility of one or another geno- 
type being unrepresented in some generation in a small population — 
have been reported for endogamous castes in India, Jews in the Ghetto 
of Eome, and small tribal groups in several areas of the world. 

In breeding populations of every type, however, there are always some 
immigrants who come, join the group (or reside with it) and have 
children. They would tend to reintroduce genes accidentally lost through 
the effect in small populations of the vagaries of mate selection and 
differing sizes of families. In the case of the Dunkard community 
immigration has been at the rate of some 10 to 22% per generation, but 
even this high rate did not prevent significant random fluctuations in 
gene frequencies. Aside from island populations such as the cases cited 
above, Buzios Island off the coast of Brazil (Williams, 1952), Puka 
Puka in the Pacific (Davies, personal communication), and a few others, 
reports of extreme reproductive isolation are often exaggerated. Authentic 
evidence for small breeding populations with low immigration rates is 
rare. A moderate immigration rate is more common. For instance, 
among unacculturated Australian aborigines a careful study indicates 
approximately 15% of matings to be intertribal (Tindale, 1953). Quite 
possibly very small population size leads to disappearance of the group 
either through fusion with another group or through extinction. Thus a 


number of small American Indian tribes have lost their distinctness as 
breeding isolates through fusion with other groups. 

As a rule of thumb one can say that appreciable random genetic drift 
will occur when the coefficient of breeding isolation (the product of the 
effective immigration rate by the effective size of the breeding population) 
is less than 50 and especially when it is less than five. Even so, such 
conditions would have to continue for a number of generations without 
marked increases in the immigration rates. Eandom genetic drift would 
be particularly important if there were some generations with a sharply 
reduced number of parents. In fact, contemporary local groups are not 
usually sharply different from others nearby and good examples of the 
effects of isolation are rare today even in relatively primitive societies; 
quite possibly conditions leading to random genetic drift have always 
been rare in human groups. 

Anthropometric measurements are not a good index of genetic homo- 
zygosity, but two studies of such measurements are instructive in this 
connection. In one, subgroups from each of two related populations on 
Bougainville Island were measured (Oliver and Howells, 1957). Among 
the Siuai there are more intergroup marriages than among the Nagovisi. 
Compared to the Siuai the Nagovisi show more intra-subgroup homo- 
geneity but not the greater inter-subgroup differences which the hypoth- 
esis of random genetic drift would require. In the other study, in three 
Peruvian towns with coefficients of isolation ranging from 100 to 300, 
clear evidence exists of increased variability among offspring of immi- 
grants, and there are some mean differences in measurements between the 
communities (Lasker, in preparation). However, such differences reflect 
separate origins or diverse proportions of Spanish, Indian and other 
elements, and one can only say that migration has not yet completely 
erased the differences between towns. There is no evidence that random 
genetic drift has created them. 

In the discussion so far I have omitted mention of incest tabus and 
special categories of preferred matings. In general the incest tabus have 
little effect on the eventual randomness of mating within a population, 
since the bans rarely extend bilaterally for more than a few generations. 
Preferred mating with kinsmen such as first cousins is more likely to 
produce a genetic situation significantly different from that in which 
random breeding within the population is the rule. Any such departure 
from randomness reduces the effective size of the breeding circle, but 
it is also rarely consistent enough to produce sharp inbreeding for enough 
generations to effect a marked increase in the likelihood of random genetic 


One should not minimize genetic drifts however, merely because it is 
rarely marked or because the differences it can produce explain only 
small local developments which presumably have no adaptive significance. 
The process can go on simultaneously in large numbers of more or less 
discrete populations, any one of which could subsequently burgeon into 
a numerous people either on the basis of some advantage of its genetic 
traits or because of some unrelated factor in its history. Although the 
exact number of races is indefinite, there are, at most, only a few major 
races of man. Furthermore, adaptive advantages of most of the genetic 
differences between them are not known and must be small or nonexistent. 
Isolation, and resultant genetic drift, therefore, may well explain the 
origin of such differences. Migration, however, will not explain the 
subsequent establishment of large racial groups unless one appeals to 
high rates of population expansion. High rates of population expansion 
do exist, however. The religious sect of Hutterites grew from 443 in 
1880 to 8,542 in 1950 without either appreciable immigration or growth 
from conversion (Eaton and Mayer, 1953). The French Canadian com- 
munity has expanded from 5,000 odd when immigration virtually ended 
in the 17th century to over 4,000,000 today. 

Migration and isolation may have indirect as well as direct effects. 
Indirect effects involve those factors which influence conditions for 
natural selection, such as new physical environments, including changed 
climates, different altitudes, and altered composition of drinking water, 
atmosphere and soil. Human migrants may also indirectly influence 
natural selection when the cultural traits they introduce alter the selec- 
tion pressures. Different marriage customs, food habits, modes of liveli- 
hood, medical practices, size and type of social groups, etc., can all serve 
to alter the relative selective values of genetic traits. 

There is also the possibility of interaction between the direct and 
indirect effects of migration. Individuals of various genotypes differ in 
their mode of response to changed cultural and natural environments in 
such a way as to produce different alterations in gene frequency. Such 
interaction is therefore evolutionary in the biogenetic sense. 

In his contribution to this symposium Hulse has already discussed 
human plasticity as manifest in studies of human migrants. He argues 
that plasticity as such is a useful characteristic in a mobile species such 
as man. Hulse suggests that this will select for certain genotypes, 
perhaps heterozygotes at numerous loci. There has sometimes been 
confusion between plastic changes in the life of the migrant individual 


and genetic adaptive changes in his descendants. It is sometimes 
assumed that adaptive changes will take the same direction and follow 
the impetus of the plastic changes in the individual. Indeed this may 
be true in situations of extreme environmental stress. Certain fruit flies 
develop a double thorax in an atmosphere with ether, and this apparently 
permits survival, while descendants who acquire the trait at earlier ages 
and transmit it genetically are evolved by selection from the stock. 

Under conditions of migration with less environmental shock, how- 
ever, plasticity would tend to relieve the group of the necessity of 
evolving specialized hereditary capacities to meet each environmental 
problem. Hence it slows down shifts in gene frequency toward specific 
environmental adaptations. Under improved dietary conditions migrants 
grow somewhat larger. Larger individuals presumably require more 
calories to maintain life. On return of adverse conditions of food supply 
the children of large plastic individuals will have their growth inhibited 
again, and these changes can all occur without change in genotype and 
without the delays and risks of extinction which genetic selection requires. 

In summary let me repeat that isolation through random genetic 
drift, though conditions for it are now rare, may be responsible for some 
of the local racial differences in man. Marked increase in numbers in a 
few small breeding populations occasionally occurs without a genetic 
swamping by immigrants and may account even for major racial groups. 
But natural selection also operates locally on local conditions in rela- 
tively isolated groups. In any case, interbreeding and migration in the 
genetically effective sense must account for the universal presence in man 
of his distinctive adaptive evolutionary advances. That is, there is always 
enough interbreeding, at least among nearby communities, to ensure that 
our descendants can receive w ithin a few ce nturies any beneficial mutant 
no matter in what place or race it originally arises. Similarly, through 
migration and intermating our adaptations are the potential heritage of 
any future human group that needs them. 


Davies, G. N. Personal communication. 

Eaton, J. W., and A. J. Mayer 1953 The social biology of very high fertility 

among the Hutterites. Human Biol., 25 : 206-264. 
Glass, Bentley 1956 On the evidence of random genetic drift in human 

populations. Am. J. Phys. Anthrop., 14: 541-555. 
Glass, Bentley, M. S. Sacks, Elsa F. John, and Charles Hess 1952 Genetic 


drift in a religious isolate: an analysis of the cause of variation in blood 
group and other gene frequencies in a small population. Am. Nat., 86: 

Kuczynski, R. R. 1953 Demographic survey of the British Colonial Empire. 
Oxford University Press, London, vol. 3. 

Lasker, G. W. 1954 Human evolution in contemporary communities. South- 
western J. Anthrop., 10 : 353-365. 

Oliver, Douglas, and W. W. Howells 1957 Micro-evolution: cultural elements 
in physical variation. Am. Anthrop., 59: 965-978. 

Shapiro, Harry L. 1929 Descendants of the Mutineers of the Bounty. Mem. 
Bernice P. Bishop Museum, vol. 11, no. 1. Honolulu, 1929. 

Tindale, Norman B. 1953 Tribal and intertribal marriage among the Aus- 
tralian Aborigines. Human Biol., 25: 169-190. 

Williams, E. 1952 Buzios Island, a Caigara Community in Southern Brazil. 
Monogr. Am. Ethnol. Soc, No. 20, J. J. Augustin, Locust Valley, New York. 


P. 0. Bow 5199, Honolulu, Hawaii 


AS is often the case when attempts are made to bridge two diverse 
J^\^ aspects of knowledge, the chairman of a symposium such as this 
is faced with a problem: shall he obtain a specialist in formal evolutionary 
theory, who may perhaps fail to devote a proper share of his attention 
to ionizing radiation, or shall he invite, say, a radiation physicist, who 
dimly recalls having heard something about evolution when he was in 
high school? Or should he ask a geneticist, who might just possibly 
equate a minute increase in radiation with the extinction of humanity? 
Or a radiologist, who will sternly remind us that radiation is an indis- 
pensable boon to mankind? Or perhaps a gentleman from the govern- 
ment, who will cheerfully advise us to " keep smiling " ? 

Your chairman has solved his problem by bravely inviting a speaker 
who cannot qualify as an expert in any of these areas. Whether this 
is an elegant solution remains to be seen. 

There is a large literature related to ionizing radiation, 1 some of it 

1 The word " radiation," as used in the present report, does not refer to the 
process of the same name which has a secure place in the literature of evolution. 
As here used, it refers throughout to ionizing radiation or to irradiation. 

The purpose of the present report is to give a brief general survey of the 
subject, and to offer certain opinions on the problems involved. It is hoped that 
the distinction between the two objectives has been clearly made. 

Problems of human evolution, as they relate to ionizing radiation, are reached 
only by passing through a number of disciplines, with a constantly accumulating 
and overlapping literature. The selected list of referencs, through June 30, 
1957, worked up by Little (1957: 1996-2053), is very useful, as are the references 
listed in the recent United Nations report on the effects of atomic radiation 
(1958). A number of surveys (Dean, 1954; Lapp, 1956; Titterton, 1956; 
Alexander, 1957; Schubert and Lapp, 1957; Wallace and Dobzhansky, 1959), 
more or less pertinent to our interests, have been written. Pauling (1958a) and 
Teller and Latter (1958) may be considered as representing opposing points of 

Several periodicals, such as Bulletin of the Atomic Scientists (Selove and 
Elkind, eds., 1958) and Scientific American, October 1959, have given entire issues 


in official reports. 2 Much of this material is rather technical and some- 
what removed from the normal reading matter of the anthropologist. 
Nevertheless, we must bring the relevant aspects of the subject into 
our area, since data meaningful to evolutionary theory are rapidly 

However, as an act of charity, some of the references and mathematics 
have been banished to the footnotes. 

First, a preliminary paragraph may be useful. Ionizing radiation 
consists of either particles or electromagnetic waves which have enough 
energy to remove electrons from atoms or molecules, the ejected products 
being known as ions. This action can be destructive to biologic organiza- 
tion, by injuring or killing the living cell in any of a number of ways. 3 

to the subject of radiation and man. In the latter journal, articles by Beadle 
(1959), Crow (1959), Hollaender and Stapleton (1959) and Platzman (1959) are 
particularly relevant. 

2 Principal governmental and official publications, in addition to special and 
semi-annual reports by the U. S. Atomic Energy Commission, include Volume 11 
of the United Nations (1956) report on peaceful uses of atomic energy; the report 
by the British Medical Research Council (1956) on the hazards to man of nuclear 
radiations; publications by the National Academy of Sciences on pathologic 
(1956b) and biologic (1956a) effects of atomic radiation; and the report by the 
World Health Organization (1957) on the effect of radiation on human heredity. 
See also reviews by Glass (1956, 1958). 

Particularly important is the voluminous " Hearings on the Nature of Radio- 
active Fallout," held before the Joint Committee on Atomic Radiation (1957). 
The definitive report of the United Nations Scientific Committee on the Effects of 
Atomic Radiation (1958) was summarized in Science (1958). Most recent is 
the report made to the A. E. C. by the General Advisory Committee (1959), 
reviewed in Science (1959b). 

3 The roentgen is the unit of exposure dose of X- or gamma radiation. The rad 
is the unit of absorbed dose. The rem ( roentgen-equivalent for man) is the 
measurement of biological effectiveness. The table below (Gladstone, 1958: 595) 
gives relationships: 

Relationship Among Radiation Units 
type of radiation 
X-rays and Gamma rays 
Beta particles 
Fast neutrons 
Thermal neutrons 
Alpha particles 

General studies of the actions of radiation on cells are given in Hollaender 
(1954), volume 1 (2 parts) and Lea (1955). The United Nations report (1958) 
reviews information in this area. 
















The biologic effects may be somatic or genetic — and the latter area is 
our particular concern. Ionizing radiation, operating on the genetic 
mechanism, inducing possible mutations, opens the door to a considera- 
tion of evolutionary consequences. 

It is generally accepted that there is a lineal relation between gonadal 
radiation and mutation, that even the smallest dose may have genetic 
effects, that the consequences are cumulative, and that there is no recovery. 
Mutations are considered to be usually deleterious. It has been suggested 
that "man may prove to be unusually vulnerable to ionizing radiations, 
including continuous exposure at low levels, on account of his known 
sensitivity to radiation, his long life, and the long interval between 
conception and the end of the reproductive period" (United Nations, 

Having said this, we must make the startling confession that there is, 
in man, "no completely convincing evidence that mutations are induced 
by radiation" (Wallace and Dobzhansky, 1959:85). However, while 
we await the evidence, it would be wise to proceed in our thinking and 
acting as if the proof had in fact been obtained. Failure to do so might 
give us the dubious honor of locking the biggest barn on earth after 
the theft of the world's largest horse. 

Meanwhile, as Neel (1958) reminds us, this area is one of the most 
actively discussed topics in human biology; furthermore, the biological 
effects of radiation appear to be one of those unhappy subjects, wherein 
the more we learn the less we like it. It is not reassuring to read that 
DNA (the desoxyribonuclear acids), genetically the most important part of 
the chromosome, normally contains phosphorus, and that if a radioactive 
isotope of phosphorus gets into the chromosomal DNA, "the affected 
molecule is doomed" (Wallace and Dobzhansky, 1959:66). It is small 
comfort to know that in a few days the isotope of phosphorus changes 
into sulphur, which we are told has no place in DNA. 

It is a bit of a shock, having been assured that strontium-90, for all 
its dangers, is not a genetic threat, to read in the A. E. C.'s 23rd semi- 
annual report (1958, p. 413), that there may be some "minor" effect 
from its incorporation into the chromosomes themselves. " Until more is 
known about such possibilities, calculations about genetic damage must 
continue to be based on the increase of background radiation due to 
long-lived gamma-ray emitting isotopes in the fallout." We will save 
a place in our calculations for the genetic damage caused by the replace- 
ment of calcium trace elements by strontium-90. 

We have no time to pursue these intriguing topics, and must turn to 


our subject, evolutionary theory, which, as we all know, is highly objec- 
tive, completely impersonal, and deals with things and people far away 
and long ago, or yet unborn. 

Particularly, we might be interested in these questions : Has ionizing 
radiation played any role in past evolution, particularly in the formation 
of human races? What changes have taken place in man's environment 
since 1900, due to man-made ionizing radiation? What are the possible 
evolutionary consequences of such changed environment, under differing 
future circumstances? 

Before proceeding, however, we must face the fact that not only is 
knowledge as jet fragmentary and unorganized, but the nature of radia- 
tion is such that factors of politics and ethics intrude. 4 The motivation 
for certain statements on the amount and effect of radiation on man 
must be questioned. Eeports have been made which contain value judg- 
ments under the guise of scientific dicta. Bias, both conscious and uncon- 
scious, can be seen, and the speaker naturally is not immune, being just 
as prejudiced as most scientists. In these days, even ivory towers have 
separate entrances, and if I must enter either by a door marked Pauling 
or a door marked Teller, I will choose the former. 

There is reason to hope, however, that the peak of difficulties may 
have passed and that, barring a tragic reversal of recent trends, there 
will be a slow return to attitudes, motives, methods of research and 
publication of findings, which we may once again call scientific. 

past evolution: eace differences 

So far as we know, life on earth has always been exposed to small 
amounts of ionizing radiation, and considerable information exists as to 
types, amounts, distribution and possible effects of radiation from natural 
sources. 5 

4 As examples of specific incidents, see Haldane's (1955) comments on Cock- 
croft, Sturtevant's (1954) criticism of Strauss' statement of March 31, 1954, 
Lapp's (1957) analysis of the Lucky Dragon case, the anaysis by Rienow and 
Rienow (1959) of the A. E. C. program, and Mumford's (1959) criticisms. For 
an account of the " snafu " — as Strauss called it — in which Strauss admitted res- 
ponsibility for refusing to permit Muller to participate in the 1955 International 
Geneva Conference, see Muller (1955a). For a report defending the A. E. C, see 
Hughes 1957). 

5 See bibliography by Lowder and Solon (1956), and reports by Libby (1955), 
Spiers (1956), Stehney and Lucas (1956) and Schaefer (1956). The Joint Com- 
mittee Hearings (1957), Appendix J (1647-1654) give the British report on 
radiation doses from natural sources. 


The United Nations Report (1958 : 9) lists the average annual gonad 
dose to man from natural radiation as about 100 mrem, as follows: 
rocks and soil, 47; cosmic rays, 28; atmosphere, 2; internal sources, 23. 
This would total 3 rem for a 30-year span. 6 

It is not surprising that, stimulated by Muller's (1927) discovery 
that mutations can be induced by x-radiation, the possible significance 
of natural radiation to evolution should have been explored. 7 Crow 
(1959 : 160), however, appears to represent the consensus (Newell, 1956 ; 
Miller and Urey, 1959) when he concludes that "ionizing radiation is 
probably not an important factor in animal and plant evolution. If it is 
important anywhere it is probably in those species, such as man, that 
have a long life span and at least for man it is a harmful rather than a 
potentially beneficial factor." 8 

Turning to the part ionizing radiation may have had in human race 
formation, we note first the wide variability shown by various types of 
background radiation. For instance, cosmic rays increase in intensity 
with altitude and with geomagnetic latitude. Even greater variability is 
shown in the radioactivity in rocks and soils. Thus, the mean dose-rate 
for thorium in igneous rocks is 37 mrad per year, nine times that in 

In some regions, where thorium-containing sands are found, high 
levels may exist, as in Kerala, where the average was determined to be 
1270 mrad per year (United Nations, 1958: 55), with an upper limit of 
84 mrad over a 30-year period. 

Other studies still in progress may give us clues. Gentry (1959), 
for instance, in a study of lj million babies born in New York State, 
is reported to have found an association between a higher incidence of 

The chief sources of background radiation are the radioactive elements 
radium, thorium ( and their decay products ) and potassium-39, found in the crust 
of the earth. Extra-terrestrial radiation comes from cosmic rays and their 
secondary radiations. Internally, radioactive elements such as potassium-40 and 
carbon- 14 are contained in the body, and radon and thoron are taken in from the 

7 See, for instance, Calvin (1956), Beadle (1957) and Sagan (1957) for recent 
speculation and conclusions. 

8 These views are derived in part from research on Drosophila and on mice, 
in which it has been determined that the rate of spontaneous mutations from 
natural radiation is extremely low. Crow (1959: 140) estimates that in man less 
than 10% of mutations are due to natural radiation, but cautions that because of 
the uncertain state of our present knowledge " it cannot be ruled out that even a 
majority of human mutations owe their origin to radiation." 


malformed infants at birth and a higher radioactive content in rock 
and soil. If a causal relationship is confirmed, the implications are 
important. We know too that other variables — on which research has 
scarcely begun in man — may add to the intensity of radiation effect. 
Thus, starvation may increase radiosensitivity, and temperature may be 
related to genetic effects (United Nations, 1958:20,21). 

Obviously much more work is needed. It would be most valuable 
to have careful studies of high background radiation areas, such as in 
Kerala and Brazil. Such studies have not been made in detail, and thus 
recent reports (General Advisory Committee, 1959) stating "human 
beings have lived for generations " in such regions, are very misleading. 9 

The problems of research in these fields, requiring the search for small 
differences in large groups over long periods of time, are evident. Thus, 
the fact that Neel and Schull (1956) did not turn up a battalion of 
two-headed babies in their F t Hiroshima population should not be taken 
as evidence that the population was unaffected genetically — although 
I am sorry to say that this position has been taken, and by men who 
should know better. 

In any event, it is not too far-fetched, I think, to imagine an early 
human group, in an area of relatively high natural radiation, geo- 
graphically isolated from other human groups, racially differentiating 
through normal evolutionary processes, but with an assist from increased 
mutations due to higher levels of background radiation. Of course, these 
levels are quite small, in terms of their presumed mutagenic capabilities, 
but it is suggested only that radiation, rather than being an offstage 
voice, may have played at least a minor supporting role in the formation 
of human races. 


It is a most disturbing mathematical exercise to add an insignificant 
amount of radiation from Source A, a reassuringly low level of radiation 
from Source B, a permissible exposure from Source C, and a safe per- 
centage from Source D, to arrive at a total which in the United States 
averages about three times the mean natural rate (Wallace and Dob- 
zhansky, 1959:76). Eventually we are going to have to face the fact 

9 Lapp's (1959b) critique of the Advisory Committee report to A. E. C. con- 
cludes ". . . it would appear that the statement on radiation hazards by the 
General Advisory Committee to the Atomic Energy Commission is misleading." 
(p. 320) 


that two new radioactive contaminants, each amounting to only 5% of 
the natural background, do not cancel each other out, as we are some- 
times seemingly led to believe, but add up to 10%, and that the total 
now becomes 110 %. 10 

Working one's way through myriad attempts to evaluate differential 
radiation exposure in man, trying to equate a brick house with a wooden 
one, brown rice with white, radiography of the pelvis with that of the 
lumbar spine, tropospheric with stratospheric fallout, one is tempted to 
agree with the little old lady, who surely by now must have said, "If 
the Good Lord had intended us to be irradiated, we'd have been born 
with built-in radiation counters." 

At present, the main source of man-made radiation is from medical 
uses, principally diagnostic x-rays. 11 In x-ray-using countries, the mean 
annual genetically significant dose is about 100 mrem, thus doubling 
the background rate. In the United States, the figure is about half again 
as much. The use of x-rays will spread, and at present we can only hope 
that self -policing medical policies will keep exposure down. 

Eadioactive wastes from atomic energy plants loom as the greatest 
peacetime radiation threat. At the moment, there is "no general popu- 
lation hazard" (United Nations, 1958 : 15), but present disposal through 

10 Beyond this, we may have to change our concept of background radiation 
itself, in which man has evolved, from that of a benign and inert standard, to an 
active and possibly damaging source of radiation to man (United Nations, 1958: 
36). See also Crow's testimony at Joint Committee Hearings (1957: 1013), and 
Lewis (1957) on relation of leukemia to natural background. Natural background 
is a base against which additional radiation contamination can be measured, but 
not necessarily a " safe " level of radiation. 

11 Varieties of radiation hazard to man in the United States are listed by the 
U. S. Public Health Service in the Joint Committee Hearings (1957: 477-479), 
together with a bibliography. Exposure from occupational hazard, not mentioned 
in the text, is at present low, on a population basis, although individual exposures 
may be high, as in airplane pilots or uranium miners (Alexander, 1957), and 
individuals who work with radiation-emitting machines, who may receive as much 
as 1 r per hour (Braestrup and Mooney, 1959: 1074). Occupational hazards will 
rise as the use of ionizing radiation in industry, medical work and research 
expands. For an idea of what this expansion involves (with a complete omission 
of any mention of radiation hazards) see Hafstad (1957). 

As examples of the wide range of individual problems, taken from a large 
collection, see Haybittle (1958) on problems of luminous watches; Fritsch, et al. 
(1958) on radiation in French spas; Weiss and Shipman (1957) on concentration 
of cobalt-60 in killer clams, due to contamination of sea water. For a discussion 
of the concentration of low-level contamination in plankton see Rienow and 
Rienow (1959: 44). 


smoke, burial and dumping at sea may only postpone a crisis. Regu- 
lation depends on agreements at the national and international level. 
Judging from the past, when economic exploitation clashed with forces 
of conservation, the prognosis is not good. 12 

Millions of words have been written on the problem of fallout, and 
I will add only a hundred or so more. Relatively, the amount of 
radiation released by fallout from nuclear weapons tests is small, but, 
like natural radiation, it is world-wide. Its principal threat lies in 
the continuation of testing programs, and the entry of new nations into 
the nuclear club. Above all, the chief threat to mankind stems from the 
purpose of testing, which is to perfect nuclear weapons for the mass 
killing of man. 

A vast amount of research has been done on the effects of these 
"nuclear events," as they are somewhat delicately called. 13 Reports, 
however, are scattered, uncoordinated and difficult to interpret. Some 
information has not been available to independent scientists. Since 1946, 
the A. E. C. has spent 125 million dollars on biomedical investigations 
on radiation, including fallout. According to Science (1959a:1210), 
"no comprehensive up-to-date account of all the A. E. C. radiation control 
work is available." 

The reports which are available are not reassuring. For instance, 
carbon-14 (Totter, Zelle and Hollister, 1958), with its half -life of about 
5600 years, to which Pauling (1958b) has called particular attention, 
should interest any student of evolution. Carbon-14 may be a greater 
genetic hazard than any other isotope, even the better publicized caesium- 

12 Glass (1957: 245) says: "It is stated on good authority (Anderson, et al., 
1957) that a 100 megawatt heat reactor will produce annually the same quantity 
of long-lived fission products as the detonation of a 1-megaton fission bomb." 
See also DuShane, in an editorial (1957) in Science. The New York Times 
(1959a) cites the serious nature of high-level radioactive waste disposal problems, 
pointing out that with an underground tank capacity of 110 million gallons, 65 
million gallons have already been used. 

13 The tabulation of "nuclear events" is given as 173,760,000 tons of TNT 
equivalent, about one-half being fission (New York Times, 1959b). 

For a general survey of effects of fallout, see Pirie (1957), the lengthy testi- 
mony in the Joint Committee Hearings (1957) and Lapp (1957). Japanese 
research has been reported in a detailed fashion in the publications of the Japan 
Society for the Promotion of Science ( 1956) . See also the recent report on fallout 

(United States, 1959), and Lapps' (1959a) resume of this report. Snyder has 
given a reasoned position on testing programs (1957). For a recent report 
contending that fallout problems have been too strongly emphasized, see Morgan 



137. And again, at the risk of being repetitive, it must be said that the 
two contaminants — and others not mentioned — are additive in their 
effects; they do not cancel each other out. 

The testing of nuclear weapons involves ethical, legal and political 
questions which transcend national policies. For the moment testing 
has ceased, pending negotiations. The sentiment of the world is, I 
believe, in favor of a continuation of this cessation. 14 

As we wander through the wilderness of radioactive contamination 
we become lost in a forest of luminous watch dials, shoe-fitting gadgets, 
television sets, electron microscopes, radioactive killer clams, high voltage 
rectifiers, hot plankton, atomic submarines and prison " Inspectoscopes." 
I wonder if the public is really aware of the amount of radioactivity 
man lives with. 

This we do know: At the moment our wallets are bulging with 
atomic credit cards. But, if we are to believe the geneticists, in due 
time the bill will be presented, and not one radiation-induced mutation 
will be omitted. For the debt, whatever it may be, must be paid; on 
this point there appears to be little argument. 15 Even the proponents 
of continued nuclear testing have shifted their position from an earlier 
claim that there was no danger, to the grounds that the end justifies the 

Diligent and conscientious efforts have been made to quantify this 
debt. 16 The results, hypothetical though they are, and admittedly based 

14 Recent evidence of public opinion may be seen in the Gallup (1059) poll 
reports, in which 77% favored a continuation of the present ban on H-bomb tests. 
The recent actions of the United Nations, the unanimous Senate Resolution, and 
the many recent statements by the State Department and the President, are evi- 
dence of a marked change in public opinion since the summer of 1958. This change 
came about through the release to the public of official information on the dangers 
of radiation, through such publications as the United Nations report (1958). 

15 For general surveys, see Dobzhansky (1955) and Auerbach (1956). Carter 
(1956) and Muller (1956) both report on the genetic problem in man in the 
United Nations (1956) report on peaceful uses of atomic energy. See also 
Bulletin of the Atomic Scientists, editorial, (1955), and Glass (1957) and Dunn 
(1957). Muller, since 1927, has repeatedly warned of genetic dangers to man 
from excessive radiation (1950, 1955b, 1957). 

Surveys of human populations are tabulated in the United Nations report 
(1957: 195). Problems of human evolution and genetics have been discussed by 
Strandskov (1950), Kraus and White (1956), Dobzhansky and Allan (1956), 
Oliver and Howells (1957), Garn (1957) and Hunt (1959). 

16 Recent examples of attempts at quantification are Inglis (1958) on radiation 
dosage from future weapons tests under three differing circumstances; Buck 


on sketchy evidence, nevertheless provide us with a framework for our 
thinking. For instance, it is useful to know that a permanent doubling 
of man's mutation rate might increase the 4% of children born with 
important detectable genetic defects to between 5 and 8% (United 
.Nations, 1958:32). It is more than a clever exercise in mathematics 
to estimate, after trying to reconcile the imponderables, that each rad 
of additional exposure per generation could cause as many as 10 million 
affected persons, at equilibrium (p. 33). 

Many of the estimates attempt to grapple with possible future con- 
ditions. 17 A permanent world-wide exposure of 10 rad appears to be 
the most popular estimate, assuming there is no nuclear war. Beadle 
(1959:232) is more optimistic, feeling an average of less than 1 r above 
natural background could be achieved. At the opposite pole, Braunbek 
(1959) postulates a steadily increasing contamination, century after 
century, ending in the gradual extinction of mankind. 

On the whole, I think the Saturday Review (1959a: 73) offers man- 
kind a better deal: a choice of extinction, survival in small numbers, or 
being changed "beyond our recognition." There are many who feel the 
last possibility would be a definite improvement. 

But all predictions falter before the prospect of nuclear war. 18 The 

(1959) on the population size required for investigation of threshold in radiation- 
induced leukemia; and Muller and Meyer (1959) on incidence of "invisible" 
detrimental mutations. The United Nations report brings together significant 
research in this area (p. 32). In the Joint Committee Report (1957), the 
National Academy of Sciences presents a discussion of radiation hazards and 
estimates (1831-1847), including the famous dictum: "keep the dose as low as 
you can." (1845) . 

17 Wallace and Dobzhansky (1959) have recently calculated mutation rates in 
man at various levels of radiation under differing circumstances, for varying 
numbers of generations. For comments on A. E. C. estimates, see Rienow and 
Rienow (1959: 158). 

18 The most recent estimate is in Joint Committee Report (1959), on effects 
of nuclear war, issued August 31, 1959, reported in Science (1959c: 696). See also 
Saturday Review (1959b), based on this report, and Lapp's (1959c) article. An 
earlier government report (United States, 1950) is based on the Hiroshima and 
Nagasaki bombs. Kissinger (1957) discusses the possibility of " limited " nuclear 
war. The enormous literature on nuclear war, much of it having relevance in any 
consideration of man's future, ranges from Cousins' early (1945) and eloquent 
declaration that " modern man is obsolete " to such recent analyses as Mills' 
(1958) on possible causes of World War III. See also Schweitzer (1957, 1958) 
and Russell ( 1959 ) . 

General Gavin's remarks on the direction of the wind are found in Hearings 


recent Joint Committee meetings (1959) concluded that "a nuclear 
war might result in a doubling of the deleterious genes the human race 
already possesses" (Lapp, 1959c: 341). ISTeel (1959), assuming a con- 
stant post-attack population of 40 million in the United States, estimated 
defective births over the next 1000 years between 17 million and 1.2 
billion. But each estimate literally depends, as General Gavin bluntly 
said, " on which way the wind blows " at the time of the attack. 

We may be stimulated to consider a few of the factors involved. 
There would of course be the direct and traumatic genetic effect, and its 
evolutionary consequences, whatever they may be. Part of the process, 
its importance impossible to assay, would be the pressure of conscious 
selection, operating on two levels : first, as we now see it in Hiroshima, 
where identification as a survivor of the bomb appears to be a factor in 
the marriage market; and second, at the international level, since it is 
certain that prompt action would be taken by the unbombed portions of 
the world — if such remained — to set up barriers against the mass entry 
of genetically damaged victims. The protection of mankind might 
necessitate either physical segregation or rigid prohibition against pro- 
creation. Mankind can absorb, perhaps, Hiroshima and Nagasaki; the 
involvement of the survivors from the northern hemisphere wonld be a 
vastly different problem. 

We now live in an age where man has the power to alter instantly 
the evolutionary prospects of his species. Moreover, the capable and 
incapable, the adaptable and the unadaptable, the potential evolutionary 
success and the potential evolutionary failure, can all become extinct 
together in an instant. It might be difficult for man to eliminate man 
entirely, but being the patriotic and determined animal he is, he would 
give it a good try. 19 


Speaking strictly in terms of direct relation of radiation to evolution, 
one might think as follows: Over a very long time, man has evolved 
through evolutionary processes now fairly well understood, which include 
certain relations between mutation and selection. These mutations were 

before Senate Committee on Armed Services in 1956. See also Bulletin of Atomic 
Scientists, Sept. 195G, p. 270. Neel's estimates were reported in Newsweek for 
July 6, 1959, p. 17. 

10 If war, as has been claimed, has been an important factor in human evolu- 
tion, the addition of the new element — radiation effects — will cause a drastic 
revision of theory. 


derived at least partly from background radiation. Presumably we 
receive a sufficiently steady supply of mutations from our background, 
to take care of future evolution, without calling in additional mutagenic 
processes, not just from man-made radiation, which is only a part of the 
picture, but from such other sources as smog, food preservatives, drugs 
and cranberry insecticides. 

To me, therefore, anything beyond the background radiation seems 
superfluous, from an evolutionary viewpoint. Excess mutations are 
functionally unnecessary, probably harmful, and possibly acutely danger- 
ous. Using as standards the characteristics suggested by the United 
Nations report (1958) as likely consequences of man-made radiation — 
excess defects at birth, smaller birth weight and stature, reduced intelli- 
gence, shortened lifespan, reduction in ability to survive or reproduce — 
I would cast my vote with those who have said, " Keep the dose as low as 
possible" (National Acadenry, 1956a). 

I can imagine man as adapting to all these factors which we usually 
consider as defects; I can imagine circumstances under which these 
things might be considered as advantages, and an entity known as man 
continue to exist. I can imagine it, but I would rather not. 

I think that man, as a species, could absorb quite a bit of radiation 
punishment, from a constant level of 10 rad to the holocaust of nuclear 
war. I can imagine man adapting himself, not to a 4% incidence of 
birth defects, but to a load of 40%. It would not be the kind of world 
we know, and I wouldn't want to live in it, but I can imagine it. 

I can imagine humanoid creatures, banded in small groups, chewing 
the bark off trees — the inner, less radioactive bark. I can imagine an 
anthropologist, in that distant time, diligently scratching in the dirt, 
trying to work out the precise level of radiation at which man may be 
said to have ceased to be human. 

I can imagine extreme adaptation to an atomically vicious environ- 
ment, with man, anthropologically recognizable, still existing. To me, 
this isn't quite the point. To me the question is, not whether man can 
adapt, but must he? Does he want to? Are his present goals so 
important, his contemporary prizes so valuable, that he will mortgage 
his evolutionary future to obtain them? 

Again, I can understand man dying for an ideal — or killing for it. 
This is a distinguishing human trait. My problem is this: aeons from 
now, when present nationalistic and ideological differences are but foot- 
notes in world history, will our successors bless us or curse us for 


defending with H-bombs what we now so positively claim to be their 
interests ? 

I do not think these are frivolous questions. We should look for 
answers, before we have so committed ourselves, by action or inaction, 
that our range of adaptation is severely limited. Since man can direct 
his own destiny — or likes to think he can — it is important that he learn 
as far as possible the alternatives. 20 

Of course, it may be that even now it is too late; that the forces 
leading inevitably to man's extinction have already been set in motion. 
It may be that this caterwauling creature, man, has at last goaded the 
cosmos into stepping on him. But even if this were so, we would still 
ask our questions, striving to derive what satisfactions we could from 
understanding the processes by which man fell even though we could 
not prevent the fall. 


I prefer to believe that it is not too late, and that man can do 
something. I suggest three areas — within the framework of our topic, 
ionizing radiation — wherein we can better prepare ourselves to answer 
problems of man's evolution. 

1. We must educate ourselves on the facts of radiation. This means 
education unhampered, uncensored and unexpurgated. Man now knows 
about this invisible and sensorially undetectable force. He fears it, but 
he has been given no chance to learn about it. Now, after a political 
detour, he can begin to learn. 21 

20 Recently a scientist high in government circles (Warren, 1959) announced 
that radiation will increase and man must learn to live with it (see also Warren 
1957). He held out the hope that medical science will curb the bad effects, and 
that man himself might ultimately build defenses against radiation. This point 
of view hardly seems to come to grips with the problem. It is less a question of 
learning to live with it, than of living to learn about it. 

21 This learning has now begun to reach the grass-roots level, as witness the 
recent articles on fallout in the Saturday Evening Post (August 29, September 5, 
1959) and Redbook (November, 1959). Some magazines have, of course, followed 
policy rather than unbiased news reporting throughout the period of debate on 
problems of radiation. See, for instance, the U. S. News (1955: 46-48) report on 
Hiroshima, claiming no A-bomb effects. 

At the national level, it is encouraging to learn that a national tabulation of 
malformations at birth will be started in 1960 by the Office of Vital Statistics. 
This program, it is announced, resulted mainly from the wide interest which has 
been shown in the hereditary effects of radioactivity. 


The pressure of honest education will inevitably force a saner use of 
radiation for man's benefit. Non-radioactive techniques to achieve the 
same results in medicine will be devised. For example, from yesterday's 
newspaper: the use of soundwaves, rather than x-rays, to detect tissue 
changes in cancer diagnosis, is being perfected. We are told by the press 
that the photographs "look like a series of jumbled white blips on a 
black surface " — which certainly sounds scientific enough. 

2. We must accept the responsibilities of a nuclear age. At a national 
level, this may be difficult, because as Beadle (1959:231) cogently 
remarks : "A nation accused of such contamination is naturally reluctant 
to face the issue squarely. The temptation is rather to avoid the issue 
by alleging that the harm does not exist or that it is insignificantly 
small. Once significant harm is admitted, the issue can no longer be 
evaded, for morality is qualitative, not quantitative." 

Accepting our responsibility, we must try to make amends to hu- 
manity. Recently a commissioner of an agency described how methods 
are being developed to lessen damage to the human system from stron- 
tium-90. At a news conference he jauntily predicted how this treatment 
would be given to us — the "us" referring to U. S. citizens. I can 
imagine nothing worse than the usurping of such a discovery for the 
exclusive use — or even the first use — of a nuclear nation. The first 
recipients of such a boon should be the innocent bystanders in non- 
nuclear nations. This would not be a favor granted, but the payment 
of an installment on a debt incurred against humanity. 

A portion of man's responsibilities include cooperation on a world- 
wide scale, through such negotiations as are now going on in Geneva, 
through the wider use of international agencies, and through conferences 
such as those on the peaceful uses of atomic energy — with the elimination, 
of course, of political maneuvering. 

It is no treason, I hope, to suggest that just as every individual is 
a part of a community, and must sacrifice some individual liberty for 
the benefits of that association, so too each nation is a part of a world 
community, and must likewise give up certain pre-atomic privileges. 22 
Which brings us to the third point. 

22 One of the most important privileges the United States may have to give up 
is that of continuing to defend the thesis that the bombing of Hiroshima was 
" necessary." For debate on this most touchy of subjects see the contrasting 
views of Blackett (1949) and Compton (1956). Jungk (1958), Amrine (1959) 
and Laurence (1959) have written recent accounts of the building and use of the 
first A-bombs, which bear on this problem. 


3. We must remove temptation from ourselves. Man must do away 
with his nuclear weapons, or face the eternal prospect of a nuclear war. 
If the first idea is impossible, and the second unthinkable, then man- 
kind has no recourse but the escape of the pathologically thwarted rat — 
insanity. I would assume that man, a creature whose neural evolution 
has developed to an exquisitely delicate degree, would be highly vul- 
nerable, in an evolutionary sense, to the long-continued effects of a world- 
wide fear neurosis. The wider context of being scared to death is being 
frightened into extinction. 

These are my three suggestions. They are not particularly inspired, 
but neither, I believe, are they unimportant. There is one more point. 

Tatum (1959:1714), when receiving the Nobel award, said: "Selec- 
tion, survival and evolution take place in response to environmental 
pressure of all kinds, including social and intellectual. In the larger 
view, the dangerous and often poorly understood and poorly controlled 
forces of modern civilization, including atomic energy and its attendant 
hazards, are but more complex and more sophisticated environmental 
challenges of life. If man cannot meet these challenges, in a biological 
sense he is not fit to survive." 

These words are clear enough, but I would go a bit further. Survival 
in a biological sense alone is not enough for man. Survival as a " victor " 
in a nuclear war is not enough. Survival as a genetically crippled monu- 
ment to an ancestral history of stupidity and greed is not enough. 

Survival in a world in which the best a man can leave his son is an 
atomic power plant, a higher probability of producing a deformed baby, 
and a nuclear pistol for atomizing fellow humans who happen to dis- 
agree with him, is, to my mind, not enough. 

In our evolution, we have acquired a few things we do not share 
with our fellow creatures. For instance, man appears to have a vague 
but universal desire to leave the world what he would call a "better 
place " in which his children can live — or to die believing he has done so. 
Today I think a generation exists which dimly suspects that, in spite 
of all our gadgets, radioactive and otherwise, we are not leaving to our 
children a better world. More than that, our children, in spite of slick, 
high-pressure atomic salesmanship, also suspect it. 

It is time, I suggest, for us as anthropologists to take a sober look 
at what man is letting himself in for, thanks to an ionized atom. 



Alexander, Peter 1957 Atomic Radiation and Life. Penguin Books, London. 
Amrine, Michael 1950 The Great Decision. G. P. Putnam's Sons, New York. 
Anderson, E. C. et al. 1957 Radioactivity and people. Science, 125: 1273-1278. 
Auerbach, Charlotte 1956 Genetics in the Atomic Age. Essential Books, 

New York. 
Beadle, G. W. 1957 Uniqueness of man. Eugenics Quart., 4: 24-29. 

1959 Ionizing radiation and the citizen. Scientific Am., 201 : 219-232. 

Blackett, P. M. S. 1949 Fear, War and the Bomb. Whittlesey House, New 

Braestrup, Carl B., and Richard T. Mooney 1959 X-ray emission from tele- 
vision sets. Science, 130: 1071-1074. 
Braunbek, Werner 1959 The Pursuit of the Atom. Emerson Books, New York. 
Buck, Carol 1959 Population size required for investigating threshold dose in 

radiation-induced leukemia. Science, 130: 1357-1358. 
Bulletin of the Atomic Sciences 1955 Editorial, and various articles, 11 : 

314-343, " Genetics and the Atom." 
Calvin, Melvin 1956 Chemical evolution and the origin of life. Am. Scientist, 

U-- 248-263. 
Carter, T. C. 1956 The genetic problem of irradiated human populations. 

United Nations Reports, 11 : 382-386. 
Compton, A. H. 1956 Atomic Quest. Oxford University Press, New York. 
Cousins, Norman 1945 Modern Man is Obsolete. Viking Press, New York. 
Crow, James F. 1959 Ionizing radiation and evolution. Scientific Am., 201 : 

Dean, Gordon 1954 Report on the Atom. Alfred A. Knopf, New York. 
Dobzhansky, Theodosius 1955 Evolution, Genetics and Man. John Wiley and 

Sons, New York. 
Dobzhansky, Theodosius, and Gordon Allan 1956 Does natural selection con- 
tinue to operate in modern mankind? Am. Anthrop., 58: 591-604. 
Dunn, L. C. 1957 Radiation and genetics. Sci. Monthy, 84'- 6-10. 
DuShane, Graham 1957 Loaded dice. Science, 125: 963. 
Fritsch, A., J. Pinset-Harstrom and J. Coursaget 1958 The concentration 

of radioactivity in water and air from hot springs at Bourboule, Mont-Dor^ 

and Royat. J. Radiol, and Electrol., 39: 148-151. 
Gallup, George 1959 Public Opinion Poll on Continuation of H-Bomb Tests. 

Star-Bulletin (Honolulu, Hawaii), December 10, 1959. 
Garn, Stanley 1957 Race and evolution. Am. Anthrop., 59: 218-224. 
General Advisory Committee 1959 Report to Joint Committee on Atomic 

Energy, made public May 7, 1959. Washington. 
Gentry, John 1959 Reported in New York Times, April 18, 1959, p. 24c. 
Gladstone, Samuel 1958 Source Book on Atomic Energy. D. Van Nostrand 

Co., New York. 
Glass, Bentley 1956 The hazards of atomic radiations to man. J. Hered., 

47: 260-268. 


1957 The genetic hazards of nuclear radiation. Science, 126: 241-246. 

1958 Effect of radiation on human heredity. Science, 128: 999. 

Hafstad, L. R. 1957 Uses of atomic radiation and energy. Sci. Monthly, 84 : 


Haldane, J. B. S. 1955 Genetical effects of radiation from products of nuclear 
explosions. Nature, 116: 115. 

Haybittle, J. L. 1958 Radiation hazards from luminous watches. Nature, 
181 : 1422. 

Hollaender, Alexandee 1954 Radiation Biology. McGraw-Hill Co., New York. 

Hollaender, Alexander, and G. E. Stapleton 1959 Radiation and the cell. 
Scientif. Am., 201 : 84-93. 

Hughes, D. J. 1957 On Nuclear Energy: its potential for peacetime use. 
Harvard University Press, Cambridge, Massachusetts. 

Hunt, E. J. 1959 Anthropometry, genetics and racial history. Am. Anthrop., 
61 : 64-87. 

Inglis, D. R. 1958 Future radiation dosage from weapon tests. Science, 127: 

Japan Society for the Promotion of Science 1956 Research in the Effects 
and Influences of the Nuclear Bomb Test Explosions. Committee for Com- 
pilation of Report on Research in the Effects of Radioactivity. Tokyo. 

Joint Committee on Atomic Energy 1957 Hearings on the Nature of Radio- 
active Fallout and its Effect on Man. Special Subcommittee on Radiation, 

1959 Report on Effects of Nuclear War. Released August 31, 1959, 


Jungk, Robert 1958 Brighter Than a Thousand Suns. Harcourt Brace and 

Co., New York. 
Kissinger, H. A. 1957 Nuclear Weapons and Foreign Policy. Harper and 

Bros., New York. 
Kraus, B. S. and C. S. White 1956 Micro-evolution in a human population. 

Am. Anthrop., 58: 1017-1043. 
Lapp, Ralph 1956 Atoms and People. Harper and Bros., New York. 

1957 The Voyage of the Lucky Dragon. Harper and Bros., New York. 

1959a Fallout hearings: second round. Bull. Atom. Scientists, 15: 


1959b A criticism of the G. A. C. report. Bull. Atom. Scientists, 15: 


1959c What is the price of a nuclear war? Bull. Atom. Scientists, 15: 


Laurence, William L. 1959 Men and Atoms. Simon and Schuster, New York. 
Lea, Douglas E. 1955 Actions of Radiations on Living Cells. Cambridge (Eng.) 
University Press. 

Lewis, E. B r 1957 Leukemia and ionizing radiation. Science, 125: 965-972. 


Libby, W. F. 1955 Dosage from natural radioactivity and cosmic rays. Science, 

122: 57-58. 
Little, Ruth A. 1957 Selected list of references. In Hearings, Joint Com- 
mittee 1957: 1996-1998, 1999-2053. 
Lowder, Wayne M., and L. R. Solon 1956 Background radiation : a literature 

search. New York, U. S. A. E. C. (NYO 4712) . 
Medical Research Council 1956 The Hazards to Man of Nuclear and Allied 

Radiations. Special Report. H. M. Stationery Office, London. 
Miller, Stanley L., and H. C. Urey 1959 Organic compound synthesis on the 

primitive earth. Science, 130: 245-251. 
Mills, C. Wright 1958 The causes of world war three. Simon and Schuster, 

New York. 
Morgan, Karl Z. 1959 Human exposure to radiation. Bull. Atom. Scientists, 

15: 384-389. 
Muller, H. J. 1927 Artificial transmutation of the gene. Science, 66: 84-87. 

1950 Our load of mutations. Am. J. Human Genet., 2: 111-176. 

1955a Comments on the genetic effects of radiation on human popula- 
tions. J. Hered., 46: 199-200. 

1955b Genetic damage produced by radiation. Science, 121 : 837-840. 

1956 How radiation changes the genetic constitution. United Nations 

Reports, 11 : 387-399. 

1957 Man's place in living nature. Sci. Monthy, 8£: 245-254. 

Muller, H. J., and Helen U. Meyer 1959 Further evidence of the relatively 

high rate of origination of " invisible " detrimental mutations. Science, 

130: 1422. 
Mumford, Lewis 1959 The morals of extermination. Atlantic Monthly, 20^: 

National Academy of Sciences 1956a The Biological Effects of Atomic 

Radiation. Washington. 

1956b Pathologic Effects of Atomic Radiation. Washington, Publica- 
tion No. 452. 

Neel, James V. 1958 The study of natural selection in primitive and civilized 
human populations. Human Biol., 30 : 43-72. 

1959 Report in Newsweek, July 6, 1959, p. 17. 

Neel, James V., and W. J. Schull 1956 The Effect of Exposure to the Atomic 
Bombs on Pregnancy Termination in Hiroshima and Nagasaki. National 
Academy of Sciences, Washington. 

Newell, Norman D. 1956 Catastrophism and the fossil record. Evolution, 10: 

New York Times 1959a Radioactive waste. September 3, p. 19: 7. 

1959b Nuclear events. August 24, p. 14. 

Oliver, Douglas, and W. W. Ho wells 1957 Micro-evolution: cultural elements 
in physical variation. Am. Anthrop., 59: 965-978. 


Pauling, Linus 1958a No More War! Dodd Mead and Co., New York. 

1958b Genetic and somatic effects of carbon- 14. Science, 128: 1183-1186. 

Pirie, A. 1957 Fallout. Greenberg, New York. 

Platzman, R. L. 1959 What is ionizing radiation? Scientific Am., 201 : 74-83. 
Rienow, Robert, and Leona T. Rienow 1959 Our New Life with the Atom. 

Thomas Y. Crowell Co., New York. 
Russell, Bertrand 1959 Common Sense and Nuclear Warfare. Simon and 

Schuster, New York. 
Sagan, Carl 1957 Radiation and the origin of the gene. Evolution, 11 : 40-55. 
Saturday Review 1959a The Research Frontier, Nov. 14, pp. 72-73. 

1959b The bombing of St. Louis, Nov. 28, pp. 15-18. 

Schaefer, Hermann J. 1956 Biological significance of the natural background 

of ionizing radiation. J. Aviation Med., 26: 453-462. 
Schubert, Jack, and R. E. Lapp 1957 Radiation. The Viking Press, New 

Schweitzer, Albert 1957 A declaration of conscience. Sat. Rev., May 18, 17- 


1958 Peace or Atomic War? Henry Holt and Co., New York. 

Science 1958 News of science. 128: 402-406. 

1959a Report on A. E. C. 129: 1210-1212. 

1959b Reports disagree on radiation hazards. In News of Science, 

129: 1473-1474. 

1959c Report on Joint Committee Report. 130: 696. 

Selove, Walter, and M. Elkind (Eds.) 1958 Radiation and man. Bull. Atom. 

Scientists, 14: No. 1. Entire issue. 

Snyder, Laurence H. 1957 What we most need to know. Sci. Monthly, 84 : 

Spiers, F. W. 1956 Radioactivity in man and his environment. Brit. J. Radiol., 

Stehney, A. F., and H. F. Lucas, Jr. 1956 Studies on the radium content of 

humans arising from the natural radium of their environment. United 

Nations Reports, 11 : 49-54. 
Strandskov, H. H. 1950 The genetics of human populations. Cold Spring 

Harbor Symposium on Quantitative Biology, 15: 1-12. 
Sturtevant, A. H. 1954 Social implications of the genetics of man. Science, 

120 : 405-407. 

Tatum, E. L. 1959 A case history in biological research. Science, 129: 1711- 

Teller, Edward, and A. L. Latter 1958 Our Nuclear Future. Criterion Books, 
New York. 

Titterton, E. W. 1956 Facing the Atomic Future. Macmillan & Co., Ltd., 


Totter, John R., M. R. Zelle, and H. Hollister 1958 Hazard to man oi 

carbon-14. Science, 128: 1490-1495. 

United Nations 1956 Peaceful Uses of Atomic Energy. Proceedings of the 

International Conference in Geneva, August, 1955. New York, United 


1958 Report on the Effects of Atomic Radiation. The United Nations 

Scientific Committee on the Effects of Atomic Radiation, New York. 

United States 1950 The Effects of Atomic Weapons. Superintendent of Docu- 
ments, U. S. Govt. Ptg. Office, Washington. 

1959 Fallout from Nuclear Weapons Tests. U. S. Govt. Ptg. Office, 

44272, Washington. 

United States Atomic Energy Commission 1958 Semi-annual report. 

U. S. News and World Report 1955 Report on Hiroshima: Thousands of 
Babies, No A-Bomb Effects. (Apr. 8) 38: 46-48. 

Wallace, Bruce, and Th. Dobzhansky 1959 Radiation, Genes and Man. 
Henry Holt, New York. 

Warren, Shields 1957 Radiation and the human body Sci. Monthly, 84: 3-6. 

■ 1959 Report in New York Times, Sept. 30, p. 22: 5. 

Weiss, H. V., and W. H. Shipman 1957 Biological concentration by killer 
clams of cobalt-60 from radioactive fallout. Science, 125: 695. 

World Health Organization 1957 The Effect of Radiation on Human Hered- 
ity. Columbia University Press, New York. 

Jacket designed by Richard Kinney 

Printed in Bulmer and Fairfield type faces 

on Warren's Olde Style Antique Wove paper 

Bound in Joanna Natulin 
Manufactured in the United States of America 

Date Due 

APR P^IQftl 




rfrn (J h WM 

L ; : : 







• *•.--*; « 





. '. 






: i , '. : -'s 

• * * ' , 

S76. i 

31 262 00066 9381 

■■.-. ••' . • " •• ,:•:■■■::